LISSAGE DE GAIN DANS UN DECODEUR DE SIGNAUX VOCAUX ET AUDIO A LARGE BANDE

GAIN-SMOOTHING IN WIDEBAND SPEECH AND AUDIO SIGNAL DECODER

Abrégé français

La présente invention concerne un procédé et un dispositif de lissage de gain modifiant l'amplitude d'un nouveau vecteur de code par rapport au bruit de fond présent dans un signal à large bande préalablement échantillonné. Ce dispositif de lissage de gain comprend un calculateur de lissage de gain servant à calculer un tel lissage en réaction à un facteur représentant la voix dans le signal à large bande échantillonné, un facteur représentant la stabilité d'un ensemble de coefficients de filtre de prédiction linéaire, et un nouveau gain de table de code. Le dispositif de lissage de gain comprend également un amplificateur servant à amplifier le nouveau vecteur de code au moyen du lissage de gain, et ce afin de produire un nouveau vecteur de code à gain lissé. La fonction du dispositif de lissage de gain améliore le signal synthétisé perçu lorsque le bruit de fond est présent dans le signal à large bande échantillonné.

Abrégé anglais

The gain smoothing method and device modify the amplitude of an innovative
codevector in relation to background
noise present in a previously sampled wideband signal. The gain smoothing
device comprises a gain smoothing calculator for
calculating a smoothing gain in response to a factor representative of voicing
in the sampled wideband signal, a factor representative of
the stability of a set of linear prediction filter coefficients, and an
innovative codebook gain. The gain smoothing device also
comprises an amplifier for amplifying the innovative codevector with the
smoothing gain to thereby produce a gain-smoothed innovative
codevector. The function of the gain-smoothing device improves the perceived
synthesized signal when background noise is present
in the sampled wideband signal.

Revendications

43
The embodiments of the invention in which an exclusive property
or privilege is claimed are defined as follows:
1. A method for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said method comprising:
finding a codevector in relation to at least one first signal encoding
parameter of said set;
calculating a first factor representative of voicing in the wideband
signal in response to at least one second signal encoding parameter of
said set;
calculating a second factor representative of stability of said
wideband signal in response to at least one third signal encoding
parameter of said set;
calculating a smoothing gain in relation to said first and second
factors; and
amplifying the found codevector with said smoothing gain to
thereby produce said gain-smoothed codevector.
2. A gain-smoothed codevector producing method as claimed in
claim 1, wherein:
finding a codevector comprises finding an innovative codevector
in an innovative codebook in relation to said at least one first signal
encoding parameter; and
the smoothing gain calculation comprises calculating the
smoothing gain also in relation to an innovative codebook gain forming a
fourth signal encoding parameter of said set.
3. A gain-smoothed codevector producing method as claimed in
claim 1, wherein:

44
finding a codevector comprises finding a codevector in a
codebook in relation to said at least one first signal encoding parameter;
and
said at least one first signal encoding parameter comprises an
innovative codebook index.
4. A gain-smoothed codevector producing method as claimed in
claim 1, wherein:
finding a codevector comprises finding an innovative codevector
in an innovative codebook in relation to said at least one first signal
encoding parameter; and
said at least one second signal encoding parameter comprises
the following parameters:
a pitch gain computed during encoding of the wideband
signal;
a pitch delay computed during encoding of the wideband
signal;
an index j of a low-pass filter selected during encoding of
the wideband signal and.applied to a pitch codevector computed
during encoding of the wideband signal; and
an innovative codebook index computed during encoding
of the wideband signal.
5. A gain-smoothed codevector producing method as claimed in
claim 1, wherein said at least one third signal encoding parameter comprises
coefficients of a linear prediction filter calculated during encoding of the
wideband signal.
6. A gain-smoothed codevector producing method as claimed in
claim 1, wherein:
finding a codevector comprises finding an innovative codevector in an
innovative codebook in relation to an index k of said innovative codebook,
said
index k forming said at least one first signal encoding parameter; and

45
calculating a first factor comprises computing a voicing factor rv by means
of the following relation:
rv = (Ev - Ec) / (Ev + Ec)
where:
- Ev is the energy of a scaled adaptive codevector bvT;
- Ec is the energy of a scaled innovative codevector gck;
- b is a pitch gain computed during encoding of the wideband
signal;
- T is a pitch delay computed during encoding of the
wideband signal;
- vT is an adaptive codebook vector at pitch delay T;
- g is an innovative codebook gain computed during
encoding of the wideband signal;
- k is an index of the innovative codebook computed during
encoding of the wideband signal; and
- ck is the innovative codevector of said innovative codebook
at index k.
7. A gain-smoothed codevector producing method as claimed in
claim 6, wherein the voicing factor rv has a value located between -1 and 1,
wherein value 1 corresponds to a pure voiced signal and value -1 corresponds
to
a pure unvoiced signals.
8. A gain-smoothed codevector producing method as claimed in
claim 7, wherein calculating a first factor comprises computing a factor
.lambda. using
the following relation:
.lambda. = 0.5 (1 - rv).

46
9. A gain-smoothed codevector producing method as claimed in
claim 6, wherein a factor .lambda.=0 indicates a pure voiced signal and a
factor .lambda.=1
indicates a pure unvoiced signal.
10. A gain-smoothed codevector producing method as claimed in
claim 1, wherein calculating a second factor comprises determining a distance
measure giving a similarity between adjacent, successive linear prediction
filters
computed during encoding of the wideband signal.
11. A gain-smoothed codevector producing method as claimed in
claim 10, wherein:
the wideband signal is sampled prior to encoding, and is
processed by frames during encoding and decoding; and
determining a distance measure comprises calculating an
Immitance Spectral Pair distance measure between the Immitance
Spectral Pairs in a present frame n of the wideband signal and the
Immitance Spectral Pairs of a past frame n-1 of the wideband signal
through the following relation:
<IMG>
where p is the order of the linear prediction filters.
12. A gain-smoothed codevector producing method as claimed in
claim 11, wherein calculating a second factor comprises mapping the Immitance
Spectral Pair distance measure Ds to said second factor .theta. through the
following
relation:

47
bounded by 0 .ltoreq. .theta. .ltoreq. 1.
13. A gain-smoothed codevector producing method as claimed in
claim 1, wherein calculating a smoothing gain comprises calculating a gain
smoothing factor Sm based on both the first .lambda. and second .theta.
factors through the
following relation:
S m=.lambda..theta.
14. A gain-smoothed codevector producing method as claimed in
claim 13, wherein the factor Sm has a value approaching 1 for an unvoiced and
stable wideband signal, and a value approaching 0 for a pure voiced wideband
signal or an unstable wideband signal.
15. A gain-smoothed codevector producing method as claimed in
claim 1, wherein:
finding a codevector comprises finding an innovative codevector
in an innovative codebook in relation to said at least one first signal
encoding parameter;
the wideband signal is sampled prior to encoding, and is
- processed by frames and subframes during encoding and decoding; and
calculating a smoothing gain comprises computing an initial
modified gain g0 by comparing an innovative codebook gain g computed
during encoding of the wideband signal to a threshold given by the initial
modified gain from the past subframe g-1 as follows:
if g<g-1 then g0=g × 1.19 bounded by g0 .ltoreq. g-1
and
if g .gtoreq. g-1 then g0=g/1.19 bounded by g0 .gtoreq. g-1.

48
16. A gain-smoothed codevector producing method as claimed in
claim 15, wherein calculating a smoothing gain comprises determining said
smoothing gain through the following relation:
g s=S m*g0+(1-S m)*g
17. A method for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said signal containing stationary background noise and said method
comprising:
finding a codevector in relation to at least one first signal encoding
parameter of said set;
calculating at least one factor indicative of stationary background
noise in the signal in response to at least one second signal encoding
parameter of said set;
calculating, in relation to said noise representative factor, a
smoothing gain using a non linear operation; and
amplifying the found codevector with said smoothing gain to
thereby produce said gain-smoothed codevector.
18. A method for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said method comprising:
finding a codevector in relation to at least one first signal encoding
parameter of said set;
calculating at least one factor indicative of stationary background
noise in the signal in response to at least one second signal encoding
parameter of said set, said at least one factor indibative of stationary
background noise comprising a factor representative of voicing in the
signal;

49
calculating, in relation to said at least one noise indicative factor
including the voicing representative factor, a smoothing gain using a non
linear operation; and
amplifying the found codevector with said smoothing gain to
thereby produce said gain-smoothed codevector.
19. A method for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said method comprising:
finding a codevector in relation to at least one first signal encoding
parameter of said set;
calculating at least one factor indicative of stationary background
noise in the signal in response to at least one second signal encoding
parameter of said set, said at least one factor indicative of stationary
background noise comprising a factor representative of stability of said
wideband signal;
calculating, in relation to said at least one noise indicative factor
including the stability representative factor, a smoothing gain using a non
linear operation; and
amplifying the found codevector with said smoothing gain to
thereby produce said gain-smoothed codevector.
20. A device for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said device comprising:
means for finding a codevector in relation to at least one first
signal encoding parameter of said set;
means for calculating a first factor representative of voicing in the
wideband signal in response to at least one second signal encoding
parameter of said set;
means for calculating a second factor representative of stability of
said wideband signal in response to at least one third signal encoding
parameter of said set;

50
means for calculating a smoothing gain in relation to said first and
second factors; and
means for amplifying the found codevector with said smoothing
gain to thereby produce said gain-smoothed codevector.
21. A gain-smoothed codevector producing device as claimed in
claim 20, wherein:
the means for finding a codevector comprises means for finding
an innovative codevector in an innovative codebook in relation to said at
least one first signal encoding parameter; and
the smoothing gain calculating means comprises means for
calculating the smoothing gain also in relation to an innovative codebook
gain forming a fourth signal encoding parameter of said set.
22. A gain-smoothed codevector producing device as claimed in
claim 20, wherein:
the means for finding a codevector comprises means for finding a
codevector in a codebook in relation to said at least one first signal
encoding parameter; and
said at least one first signal encoding parameter comprises an
innovative codebook index.
23. A gain-smoothed codevector producing device as claimed in
claim 20, wherein:
the means for finding a codevector comprises means for finding
an innovative codevector in an innovative codebook in relation to said at
least one first signal encoding parameter; and
said at least one second signal encoding parameter comprises
the following parameters:
a pitch gain computed during encoding of the wideband
signal;
a pitch delay computed during encoding of the wideband
signal;

51
an index j of a low-pass filter selected during encoding of
the wideband signal and applied to a pitch codevector computed
during encoding of the wideband signal; and
an innovative codebook index computed during encoding
of the wideband signal.
24. A gain-smoothed codevector producing device as claimed in
claim 20, wherein said at least one third signal encoding parameter comprises
coefficients of a linear prediction filter calculated during encoding of the
wideband signal.
25. A gain-smoothed codevector producing device as claimed in
claim 20, wherein:
the means for finding a codevector comprises means for finding
an innovative codevector in an innovative codebook in relation to an
index k of said innovative codebook, said index k forming said at least
one first signal encoding parameter; and
the means for calculating a first factor comprises means for
computing a voicing factor rv by means of the following relation:
rv = (Ev - Ec) / (Ev + Ec)
where:
- Ev is the energy of a scaled adaptive codevector bvT;
- Ec is the energy of a scaled innovative codevector gck;
- b is a pitch gain computed during encoding of the wideband
signal;
- T is a pitch delay computed during encoding of the
wideband signal;
- vT is an adaptive codebook vector at pitch delay T;
- g is an innovative codebook gain computed during
encoding of the wideband signal;

52
- k is an index of the innovative codebook computed during
encoding of the wideband signal; and
- ck is the innovative codevector of said innovative codebook
at index k.
26. A gain-smoothed codevector producing device as claimed in
claim 25, wherein the voicing factor rv has a value located between -1 and 1,
wherein value 1 corresponds to a pure voiced signal and value -1 corresponds
to
a pure unvoiced signals.
27. A gain-smoothed codevector producing device as claimed in
claim 26, wherein the means for calculating a first factor comprises means for
computing a factor .lambda. using the following relation: -
.lambda. = 0.5 (1 - rv).
28. A gain-smoothed codevector producing device as claimed in
claim 27, wherein a factor .lambda.=0 indicates a pure voiced signal and a
factor .lambda.=1
indicates a pure unvoiced signal.
29. A gain-smoothed codevector producing device as claimed in
claim 20, wherein the means for calculating a second factor comprises means
for
determining a distance measure giving a similarity between adjacent,
successive
linear prediction filters computed during encoding of the wideband signal.
30. A gain-smoothed codevector producing device as claimed in
claim 29, wherein:
the wideband signal is sampled prior to encoding, and is
processed by frames during encoding and decoding; and
the means for determining a distance measure comprises means
for calculating an Immitance Spectral Pair distance measure between the
Immitance Spectral Pairs in a present frame n of the wideband signal and

53
the Immitance Spectral Pairs of a past frame n-1 of the wideband signal
through the following relation:
<IMG>
where p is the order of the linear prediction filters.
31. A gain-smoothed codevector producing device as claimed in
claim 30, wherein the means for calculating a second factor comprises means
for
mapping the Immitance Spectral Pair distance measure Ds to said second factor
.theta. through the following relation:
.theta.=1.25 - D s / 400000.0
bounded by 0 .ltoreq. .theta. .ltoreq. 1.
32. A gain-smoothed codevector producing device as claimed in
claim 20, wherein the means for calculating a smoothing gain comprises means
for calculating a gain smoothing factor Sm based on both the first .lambda.
and second .theta.
factors through the following relation:
S m = .lambda..theta.
33. A gain-smoothed codevector producing device as claimed in
claim 32, wherein the factor Sm has a value approaching 1 for an unvoiced and
stable wideband signal, and a value approaching 0 for a pure voiced wideband
signal or an unstable wideband signal.
34. A gain-smoothed codevector producing device as claimed in
claim 20, wherein:

54
the means for finding a codevector comprises means for finding
an innovative codevector in an innovative codebook in relation to said at
least one first signal encoding parameter;
the wideband signal is sampled prior to encoding, and is
processed by frames and subframes during encoding and decoding; and
the means for calculating a smoothing gain comprises means for
computing an initial modified gain g O , said initial modified gain computing
means comprising means for comparing an innovative codebook gain g
computed during encoding of the wideband signal to a threshold given by
the initial modified gain from the past subframe g-1 as follows:
if g<g-1 then g O=g x 1.19 bounded by g O .ltoreq. g-1
and
if g .gtoreq. g-1 then g O=g/1.19 bounded by g O .gtoreq. g-1.
35. A gain-smoothed codevector producing method as claimed in
claim 34, wherein the means for calculating a smoothing gain comprises means
for determining said smoothing gain through the following relation:
g s=S m*g o+(1-S m)*g
36. A device for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said signal containing stationary background noise and said device
comprising:
means for finding a codevector in relation to at least one first
signal encoding parameter of said set;

55
means for calculating at least one factor indicative of stationary
background noise in the signal in response to at least one second signal
encoding parameter of said set;
means for calculating, in relation to said noise representative
factor, a smoothing gain using a non linear operation; and
means for amplifying the found codevector with said smoothing
gain to thereby produce said gain-smoothed codevector.
37. A device for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said device comprising:
means for finding a codevector in relation to at least one first
signal encoding parameter of said set;
means for calculating at least one factor indicative of stationary
background noise in the signal in response to at least one second signal
encoding parameter of said set, said at least one factor indicative of
stationary background noise comprising a factor representative of voicing
in the signal;
means for calculating, in relation to said at least one noise
indicative factor including the voicing representative factor, a smoothing
gain using a non linear operation; and
means for amplifying the found codevector with said smoothing
gain to thereby produce said gain-smoothed codevector.
38. A device for producing a gain-smoothed codevector during
decoding of an encoded wideband signal from a set of signal encoding
parameters, said device comprising:
means for finding a codevector in relation to at least one first
signal encoding parameter of said set;
means for calculating at least one factor indicative of stationary
background noise in the signal in response to at least one second signal
encoding parameter of said set, said at least one factor indicative of

56
stationary background noise comprising a factor representative of stability
of said wideband signal;
means for calculating, in relation to said at least one noise
indicative factor including the stability representative factor, a smoothing
gain using a non linear operation; and
means for amplifying the found codevector with said smoothing
gain to thereby produce said gain-smoothed codevector.
39. A cellular communication system for servicing a geographical
area divided into a plurality of cells, comprising:
mobile transmitter/receiver units;
cellular base stations respectively situated in said cells;
means for controlling communication between the cellular
base stations;
a bidirectional wireless communication sub-system between
each mobile unit situated in one cell and the cellular base station of said
one cell,
said bidirectional wireless communication sub-system comprising in both the
mobile unit and the cellular base station (a) a transmitter including a
decoder for
encoding a wideband signal and means for transmitting the encoded wideband
signal, and (b) a receiver including means for receiving a transmitted encoded
wideband signal and a decoder for decoding the received encoded wideband
signal;
wherein said decoder comprises means responsive to a set of signal
encoding parameters for decoding the received encoded wideband signal,
and wherein said wideband signal decoding means comprises a device as
recited in any one of claims 20 to 38, for producing a gain-smoothed
codevector during decoding of the encoded wideband signal from said set of
signal encoding parameters.
40. A communication network element comprising a receiver
including means for receiving a transmitted encoded wideband signal and a
decoder for decoding the received encoded wideband signal;

57
wherein said decoder comprises means responsive to a set of signal
encoding parameters for decoding the received encoded wideband signal,
and wherein said wideband signal decoding means comprises a device as
recited in any one of claims 20 to 38, for producing a gain-smoothed
codevector during decoding of the encoded wideband signal from said set of
signal encoding parameters.
41. A mobile transmitter/receiver unit comprising a receiver
including means for receiving a transmitted encoded wideband signal and a
decoder for decoding the received encoded wideband signal;
wherein said decoder comprises means responsive to a set of signal
encoding parameters for decoding the received encoded wideband
signal, and wherein said wideband signal decoding means comprises a
device as recited in any one of claims 20 to 38, for producing a gain-
smoothed codevector during decoding of the encoded wideband signal
from said set of signal encoding parameters.
42. In a cellular communication system for servicing a
geographical area divided into a plurality of cells, comprising: mobile
transmitter/receiver units; cellular base stations respectively situated in
said cells;
and means for controlling communication between the cellular base stations;
- a bidirectional wireless communication sub-system between each mobile unit
situated in one cell and the cellular base station of said one cell, said
bidirectional
wireless communication sub-system comprising in both the mobile unit and the
cellular base station (a) a transmitter including an encoder for encoding a
wideband signal and means for transmitting the encoded wideband signal, and
(b) a receiver including means for receiving a transmitted encoded wideband
signal and a decoder for decoding the received encoded wideband signal;
wherein said decoder comprises means responsive to a set of signal
encoding parameters for decoding the received encoded wideband signal,
and wherein said wideband signal decoding means comprises a device as
recited in any one of claims 20 to 38, for producing a gain-smoothed

58
codevector during decoding of the encoded wideband signal from said set of
signal encoding parameters.

Description

CA 02391562 2005-O1-07
1
GAIN-SMOOTHING METHOD AND DEVICE
FOR WIDEBAND SIGNAL ENCODER
BACKGROUND OF THE INVENTION
1. Field of the invention:
The present invention relates to a gain-smoothing method and
device implemented in a wideband signal encoder.
2. Brief description of the prior art:
The demand for efficient digital wideband speech/audio encoding
techniques with a good subjective quality/bit rate trade-off is increasing for
numerous applications such as audio/video teleconferencing, multimedia, and
wireless applications, as well as Internet and packet network applications.
Until
recently, telephone bandwidths filtered in the range 200-3400 Hz were mainly
used in speech encoding applications. However, there is an increasing demand
for wideband speech applications in order to increase the intelligibility and
naturalness of the speech signals. A bandwidth in the range 50-7000 Hz was
found sufficient for delivering a face-to-face speech quality. For audio
signals,
this range gives an acceptable audio quality, but is still lower than the CD
quality
which operates in the range 20-20000 Hz.
A speech encoder converts a speech signal into a digital bitstream
which is transmitted over a communication channel (or stored in a storage
medium). The speech signal is digitized (sampled and quantized usually with 16-
bits per sample) and the speech encoder has the role of representing these
digital samples with a smaller number ofibits while maintaining a good
subjective
speech quality. The speech decoder or synthesizer processes the transmitted or
stored bit stream to convert it back to a sound signal, for example a
speech/audio

~ .n,". F,..i.i,i~ . ~.. L.
CA 02391562 2005-O1-07
..
signal.
2
One of the best prior art techniques capable of achieving a good
quality/bit rate trade-off is the so-called Code Excited Linear Prediction
(CELP)
technique. According to this technique, the sampled speech signal is processed
in successive blocks of L samples usually called frames where L is some
predetermined number (corresponding to 10-30 ms of speech). In CELP, a linear
prediction (LP) synthesis filter is computed and transmitted every frame. The
L-
sample frame is then divided into smaller blocks called subframes of size N
samples, where L=kN and k is the number of subframes in a frame (N usually
corresponds to 4-10 ms of speech). An excitation signal is determined in each
subframe, which usually consists of two components: one from the past
excitation (also called pitch contribution or adaptive codebook) and the other
from an innovative codebook (also called fixed codebook). This excitation
signal
is transmitted and used at the decoder as the input of the LP synthesis filter
in
order to obtain a synthesized speech.
An innovative codebook in the CELP context, is an indexed set of
N-sample-long sequences which will be referred to as N-dimensional
codevectors. Each codebook sequence is indexed by an integer k ranging from 1
to M where M represents the size of the codebook often expressed as a number
of bits b; where M=2b.
To synthesize speech according to the CELP technique, each
block of N samples is synthesized by filtering an appropriate codevector from
an
innovative codebook through time varying filters modeling the spectral
characteristics of the speech signal. At the encoder end, the synthesis output
is
computed for all, or a subset, of the codevectors from the innovative codebook
(codebook search). The retained codevector is the one producing the synthesis
output closest to the original speech signal according to a perceptually
weighted
distortion measure. This perceptual weighting is performed using a so-called
perceptual weighting filter, which is usually derived from the LP synthesis
filter.
The CELP model has been very successful in encoding telephone

CA 02391562 2005-O1-07
3
band sound signals, and several CELP-based standards exist in a wide range of
applications, especially in digital cellular applications. In the telephone
band, the
sound signal is band-limited to 200-3400 Hz and sampled at 8000 samples/sec.
In wideband speech/audio applications, the sound signal is band-limited to 50-
7000 Hz and sampled at 16000 samples/sec.
Some difficulties arise when applying the telephone-band
optimized CELP model to wideband signals, and additional features need to be
added to the model in order to obtain high quality wideband signals. Wideband
signals exhibit a much wider dynamic range compared to telephone-band
signals, which results in precision problems when a fixed-point implementation
of
the algorithm is required (which is essential in wireless applications).
Furthermore, the CELP mode( will often spend most of its encoding bits on the
low-frequency region, which usually has higher energy contents, resulting in a
low-pass output signal.
A problem noted in synthesized speech signals is a reduction in
decoder performance when background noise is present in the sampled speech
signal. At the decoder end, the CELP model uses post-filtering and post-
processing techniques in order to improve the perceived synthesized signal.
These techniques need to be adapted to accomodate wideband signals.
SUMMARY OF THE INVENTION
In order to overcome the above discussed problem of the prior art,
the present invention provides a method for producing a gain-smoothed
codevector during decoding of an encoded wideband signal from a set of signal
encoding parameters. This signal contains stationary background noise and the
method comprises: finding a codevector in relation to at least one first
signal
encoding parameter of the set; calculating at least one factor indicative of
stationary background noise in the signal in response to at least one second
signal encoding parameter .of the set; calculating, in relation to the noise
representative factor, a smoothing gain using a non linear operation; and

....".... .~4,~~..,.y.,......i.
CA 02391562 2005-O1-07
4
amplifying the found codevector with the smoothing gain to thereby produce the
gain-smoothed codevector.
The present invention also relates to a method for producing a
gain-smoothed codevector during decoding of an encoded wideband signal from
a set of signal encoding parameters, this method comprising: finding a
codevector in relation to at least one first signal encoding parameter of the
set;
calculating at least one factor indicative of stationary background noise in
the
signal in response to at least one second signal encoding parameter of the
set,
the at least one factor indicative of stationary background noise comprising a
factor representative of voicing in the signal; calculating, in relation to
the at least
one noise indicative factor including the voicing representative factor, a
smoothing gain using a non linear operation; and amplifying the found
codevector with the smoothing gain to thereby produce the gain-smoothed
codevector.
The present invention further relates to a method for producing a
gain-smoothed codevector during decoding of an encoded wideband signal from
a set of signal encoding parameters, this method comprising: finding a
codevector in relation to at least one first signal encoding parameter of the
set;
calculating at least one factor indicative of stationary background noise in
the
signal in response to at least one second signal encoding parameter of the
set,
the at least one factor indicative of stationary background noise comprising a
factor representative of stability of the wideband signal; calculating, in
relation to
the at least one noise indicative factor including the stability
representative factor,
a smoothing gain using a non linear operation; and amplifying the found
codevector with the smoothing gain to thereby produce the gain-smoothed
codevector.
Still further in accordance with the invention, there is provided a
method for producing a gain-smoothed codevector during decoding of an
encoded wideband signal from a set of signal encoding parameters. This method
comprises: finding a codevector in relation to at least one first signal
encoding

. . ."" , i, ", ,~ a~, ., ,. .,
CA 02391562 2005-O1-07
parameter of the set; calculating a first factor representative of voicing in
the
wideband signal in response to at least one second signal encoding parameter
of
the set; calculating a second factor representative of stability of the
wideband
signal in response to at Least one third signal encoding parameter of the set;
5 calculating a smoothing gain in relation to the first and second factors;
and
amplifying the found codevector with the smoothing gain to thereby produce the
gain-smoothed codevector.
Accordingly, the present invention uses a gain-smoothing feature
for efficiently encoding wideband (5Q-7000 Hz) signals through, in particular
but
not exclusively, CELP-type encoding techniques, in view of obtaining high a
quality reconstructed signal (synthesized signal) especially in the presence
of
background noise in the sampled wideband signal.
In accordance with non-restrictive preferred embodiments of the
gain-smoothed codevector producing method:
- finding a codevector comprises finding an innovative codevector in an
innovative codebook in relation to the at least one first wideband signal
encoding parameter;
- the smoothing gain calculation comprises calculating the smoothing gain also
in relation to an innovative codebook gain forming a fourth wideband signal
encoding parameter of the set;
- the first wideband signal encoding parameter comprises an innovative
codebook index;
the at least one second wideband signal encoding parameter comprises the
following parameters:
a pitch gain computed during encoding of the wideband
signal;

., r .nrr~i~ ~ ~ i I' ~ir .r ylr .. .~ r., .1..
CA 02391562 2005-O1-07
6
a pitch delay computed during encoding of the wideband
signal;
an index j of a low-pass filter selected during encoding of
the wideband signal and applied to a pitch codevector computed
during encoding of the wideband signal; and
an innovative codebook index computed during encoding
of the wideband signal;
- the at least one third wideband signal encoding parameter comprises
coefficients of a linear prediction filter calculated during encoding of the
wideband signal;
- the innovative codevector is found in the innovative codebook in relation to
an index k of the innovative codebook, this index k forming the first wideband
signal encoding parameter;
- calculating a first factor comprises computing a voicing factor rv by means
of
the following relation:
rv = (Ev - Ec) / (Ev + Ec)
where:
- Ev is the energy of a scaled adaptive codevector bvT;
- Ec is the energy of a scaled innovative codevector gck;
- b is a pitch gain computed during encoding of the wideband
signal;
- T is a pitch delay computed during encoding of the
wideband signal;
- vT is an adaptive codebook vector at pitch delay T;
- g is an innovative codebook gain computed during
encoding of the wideband signal;
- k is an index of the innovative codebook computed during
encoding of the wideband signal; and

CA 02391562 2005-O1-07
7
- ck is the innovative codevector of said innovative codebook
at index k;
- the voicing factor rv has a value located between -1 and 1, wherein value
1 corresponds to a pure voiced signal and value -1 corresponds to a pure
unvoiced signals;
- calculating a first factor comprises computing a factor 7~ using the
following
relation: -
~. = 0.5 (1 - rv).
- a factor ~,=0 indicates a pure voiced signal and a factor ~,=1 indicates a
pure
unvoiced signal;
- calculating a second factor comprises determining a distance measure giving
a similarity between adjacent, successive linear prediction filters computed
during encoding of the wideband signal;
- the wideband signal is sampled prior to encoding, and is processed by
frames during encoding and decoding, and determining a distance measure
comprises calculating an Immittance Spectral Pair distance measure
between the Immitance Spectral Pairs in a present frame n of the wideband
signal and the Imrnittance Spectral Pairs of a past frame n-1 of the wideband
signal through the following relation:
p-I
D~ _ ~ (ISp; n) - ISp; n-I) ~ 2
i=1
where p is the order of the linear prediction filter;
- calculating a second factor comprises mapping the Immittance Spectral Pair
distance measure Ds to the second factor 8 through the following relation:

i~ ~ ~.~.nii~ iL,ii~~I~~Ar.. i.~.,l.
CA 02391562 2005-O1-07
8 =1. 25 - DS l 400000. 0
bounded by 0 <_ A _< 1;
- calculating a smoothing gain comprises calculating a gain smoothing factor
Sm based on both the first ~, and second 8 factors through the following
relation:
S~=~,9
- the factor Sm has a value approaching 1 for an unvoiced and stable
wideband signal, and a value approaching 0 for a pure voiced wideband
signal or an unstable wideband signal;
- calculating a smoothing gain comprises computing an initial modified gain g0
by comparing an innovative codebook gain g computed during encoding of
the wideband signal to a threshold given by the initial modified gain from the
past subframe g-1 as follows:
if g<g-1 then ,g0=g x 1.19 bounded by g0 < g-1
and
if g >_ g-1 then g0=g/1.19 bounded by g0~g-1 ; and
- calculating a smoothing gain comprises determining this smoothing gain
through the following relation:
gs=Sm*go+~1'Sm~*g

CA 02391562 2005-O1-07
9
The present invention is still concerned with:
- A device for producing a gain-smoothed codevector during decoding of an
encoded wideband signal from a set of signal encoding parameters, the
signal containing stationary background noise and the device comprising:
means for finding a codevector in relation to at least one first signal
encoding
parameter of the set; means for calculating at least one factor indicative of
stationary background noise in the signal in response to at least one second
signal encoding parameter of the set; means for calculating, in relation to
the
noise representative factor, a smoothing gain using a non linear operation;
and means for amplifying the found codevector with the smoothing gain to
thereby produce the gain-smoothed codevector.
- A device for producing a gain-smoothed codevector during decoding of an
encoded wideband signal from a set of signal encoding parameters, the
device comprising: means for finding a codevector in relation to at least one
first signal encoding parameter of the set; means for calculating at least one
factor indicative of stationary background noise in the signal in response to
at
least one second signal encoding parameter of the set, said at least one
factor indicative of stationary background noise comprising a factor
representative of voicing in the signal; means for calculating, in relation to
the
at least one noise indicative factor including the voicing~representative
factor,
a smoothing gain using a non linear operation; and means for amplifying the
found codevector with the smoothing gain to thereby produce the gain-
smoothed codevector.
- A device for producing a gain-smoothed codevector during decoding of an
encoded wideband signal from a set of signal encoding parameters, the
device comprising: means for finding a codevector in relation to at least one
first signal encoding parameter of the set; means for calculating at least one
factor indicative of stationary background noise in the signal in response to
at

. ., 4 ., .,. d" _ .. ~ " ., i..
CA 02391562 2005-O1-07
least one second signal encoding parameter of the set, said at least one
factor indicative of stationary background noise comprising a factor
representa~ve of stability of the wideband signal; means for calculating, in
relation to the at least one noise indicative factor including the stability
5 representative factor, a smoothing gain using a non linear operation; and
means for amplifying the found codevector with the smoothing gain to
thereby produce the gain-smoothed codevector.
A device for producing a gain-smoothed codevector during decoding of an
10 encoded wideband signal from a set of signal encoding parameters, the
device comprising: means for finding a codevector in relation to at least one
first signal encoding parameter of the set; means for calculating a first
factor
representative of voicing in the wideband signal in response to at least one
second signal encoding parameter of the set; means for calculating a second
factor representative of stability of the wideband signal in response to at
least
one third signal encoding parameter of:the set; means for calculating a
smoothing gain in relation to the first and second factors; and means for
amplifying the found codevector with the smoothing gain to thereby produce
the gain-smoothed codedector.
The present invention still further relates to a cellular
communication system, a communication network element, a mobile
transmitter/receiver unit, and a bidirectional wireless communication sub-
system
incorporating the above device for producing a gain-smoothed codevector during
decoding of the encoded wideband signal from the set of signal encoding
parameters.
The above and other objects, advantages and features of the
present invention will become more apparent upon reading the following non
restrictive description of a preferred embodimentthereof, given forthe purpose
of
illustration only with reference to the accompanying drawings.

~~n~~ , I, n~ "i ,i~i~ .,,. ~, .i
CA 02391562 2005-O1-07
1.1
BRIEF DESCRIPTION OF THE DRAWINGS
In the appended drawings:
Figure 1 is a schematic block diagram of a wideband encoder;
Figure 2 is a schematic block diagram of a wideband decoder embodying
gain-smoothing method and device according to the invention;
Figure 3 is a schematic block diagram of a pitch analysis device;
Figure 4 is a schematic flow chart of the gain-smoothing method
embodied in the wideband decoder of Figure 2; and
Figure 5 is a simplified, schematic block diagram of a cellular
communication system in which the wideband encoder of Figure 1 and the
wideband decoder of Figure 2 can be used.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
As well known to those of ordinary skill in the art, a cellular
communication system such as 401 (see Figure 4) provides a telecommunication
service over a large geographic area by dividing that large geographic area
into a
number C of smaller cells. The C smaller cells are serviced by respective
cellular
base stations 4021, 4022 ... 402C to provide each cell with radio signaling,
audio
and data channels.
Radio signaling channels are used to page mobile radiotelephones
(mobile transmitter/receiver units) such as 403 within the limits of the
coverage
area (cell) of the cellular base station 402, and to place calls to other
radiotelephones 403 located either inside or outside the base station's cell
or to
another network such as the Public Switched Telephone Network (PSTN) 404.

~n~~,.,.a"... ~...~,
CA 02391562 2005-O1-07
12
Once a radiotelephone 403 has successfully placed or received a call, an
audio or data channel is established between this radiotelephone 403 and the
cellular base station 402 corresponding to the cell in which the
radiotelephone
403 is situated, and communication between the base station 402 and
radiotelephone 403 is conducted over that audio or data channel. The
radiotelephone 403 may also receive control or timing information over a
signaling channel while a call is in progress.
If a radiotelephone 403 leaves a cell and enters another adjacent cell
while a call is in progress, the radiotelephone 403 hands over the call to an
available audio or data channel of the base station 402 of the new cell. If a
radiotelephone 403 leaves a cell and enters another adjacent cell while no
call is
in progress, the radiotelephone 403 sends a control message over the signaling
channel to log into the base station 402 of the new cell. In this manner
mobile
communication over a wide geographical area is possible.
The cellular communication system 401 further comprises a control
terminal 405 to control communication between the cellular base stations 402
and the PSTN 404, for example during a comi~nunication between a
radiotelephone 403 and the PSTN 404, or between a radiotelephone 403 located
in a first cell and a radiotelephone 403 situated in a second cell.
Of course, a bidirectional wireless radio communication subsystem is
required to establish an audio or data channel between a base station 402 of
one
cell and a radiotelephone 403 located in that cell. As illustrated in very
simplified
form in Figure 4, such a bidirectional wireless radio communication subsystem
typically comprises in the radiotelephone 403:
- a transmitter 406 including:
- an encoder 407 for encoding speech; and
- a transmission circuit 408 for transmitting the encoded
speech from the encoder 407 through an antenna such as 409;
and
- a receiver 410 including:
- a receiving circuit 411 for receiving transmitted encoded

CA 02391562 2005-O1-07
13
speech usually through the same antenna 409; and
- a decoder 412 for decoding the received encoded
speech from the receiving circuit 411.
The radiotelephone 403 further , comprises other conventional
radiotelephone circuits 413 to which the encoder 407 and decoder 412 are
connected and for processing signals therefrom, which circuits 413 are well
known to those of ordinary skill in the art and, accordingly, will not be
further
described in the present specification.
Also, such a bidirectional wireless radio communication subsystem
typically comprises in each base station 402:
- a transmitter 414 including:
- an encoder 415 for encoding speech; and
- a transmission circuit 416 for transmitting the encoded
speech from the encoder 415 through an antenna such as 417;
and
- a receiver 418 including:
- a receiving circuit 419 for receiving transmitted encoded
speech through the same antenna 417 or through another
antenna (not shown); and
- a decoder 420 for decoding the received encoded
speech from the receiving circuit 419.
The base station 402 further comprises, typically, a base station controller
421, along with its associated database 422, for controlling communication
between the control terminal 405 and the transmitter 414 and receiver 418.
As well known to those of ordinary skill in the art, voice encoding is
required in order to reduce the bandwidth necessary to transmit sound signals,
for example voice signal such as speech, across the bidirectional wireless
radio
communication subsystem, i.e., between a radiotelephone 403 and a base
station 402.
LP voice encoders (such as 415 and 407) typically operating at 13

. , ~"~" n, ", ,~ ,a~,.., .. ~ , ,~ i
CA 02391562 2005-O1-07
14
kbits/second and below such as Code-Excited Linear Prediction (CELP) .
encoders typically use a LP synthesis filter. to model the short-term spectral
envelope of speech. The LP information is transmitted, typically, every 10 or
20
ms to the decoder (such 420 and 412) and is extracted at the decoder end.
The novel techniques disclosed in the present specification can apply to
different LP-based encoders. However, a CELP-type encoder is used in the
preferred embodiment for the purpose of presenting a non-limitative
illustration of
these techniques. In the same manner, such techniques can be used with sound
signals other than speech and voice as well as with other types of wideband
signals.
Figure 1 shows a general block diagram of a CELP-type speech encoder
100 modified to better accommodate wideband signals.
The sampled input speech signal i 14 is divided into successive L-sample
blocks called "frames". During each frame, different parameters representing
the
speech signal in the frame are computed, encoded, and transmitted. LP
parameters representing the LP synthesis filter are usually computed once
every
frame. The frame is further divided into smaller blocks of N samples (blocks
of
length N), in which excitation parameters (pitch and innovation) are
determined.
In the CELP literature, these blocks of length N are celled "subframes" and
the N-
sample signals in the subframes are referred to as N-dimensional vectors. In
this
preferred embodiment, the length N corresponds to 5 ms while the' length L
corresponds to 20 ms, which means that a frame contains four subframes (N=80
at the sampling rate of 16 kHz and 64 after down-sampling to 12.8 kHz).
Various
N-dimensional vectors are involved in the encoding procedure. A list of
vectors
appearing in Figures 1 and 2 as well as a list of transmitted parameters are
given
herein below:
List of the main N-dimensional vectors
s Wideband signal input speech vector (after down-sampling, pre-
processing, and preemphasis);
sw Weighted speech vector;

CA 02391562 2005-O1-07
s0 Zero-input response of weighted synthesis filter;
sp Down-sampled pre-processed signal;
Oversampled synthesized speech signal;
s' Synthesis signal before deemphasis;
5 sd Deemphasized synthesis signal;
sh Synthesis signal after deemphasis and postprocessing;
x Target vector for pitch search;
x' Target vector for innovative search;
h Weighted synthesis filter impulse response;
10 vT Adaptive (pitch) codebook vector at delay
T;
yT Filtered pitch codebook vector (vT convolved
with h);
ck Innovative codevector at index k (k-th entry
from the innovative
codebook);
cf Enhanced scaled innovative codevector;
15 a Excitation signal (scaled innovative and
pitch codevectors);
u' Enhanced excitation;
z Band-pass noise sequence; .
w' White noise sequence; and
w Scaled noise sequence.
List of transmitted parameters
STP Short term prediction parameters (defining A(z));
T Pitch lag (or pitch codebook index);
b Pitch gain (or pitch codebook gain);
j Index of the low-pass filter applied to the pitch codevector;
k Codevector index (innovative codebook entry); and
g Innovative codebook gain.
In this preferred embodiment, the STP parameters are transmitted once
per frame and the rest of the parameters are transmitted four times per frame
(every subframe).

CA 02391562 2005-O1-07
16
ENCODER 100
The sampled speech signal is encoded on a block by block basis by the
encoder 100 of Figure 1 which is broken down into eleven (11 ) modules bearing
references 101 to 111, respectively.
The input speech is processed into the above mentioned L-sample blocks
called frames.
Referring to Figure 1, the sampled input speech signal 114 is down-
sampled in a down-sampling module 101. For example, the signal is down-
sampled from 16 kHz down to 12.8 kHz, using techniques well known to those of
ordinary skill in the art. Down-sampling to a frequency other than 12.8 kHz
can of
course be envisaged. Down-sampling increases the coding efficiency, since a
smaller frequency bandwidth is encoded. This also reduces the algorithmic
complexity since the number of samples in a frame is decreased. The use of
down-sampling becomes significant when the bit rate is reduced below 16
kbit/sec, although down-sampling is not essential above 16 kbit/sec.
After down-sampling, the 320-sample firame of 20 ms is reduced to a 256-
sample frame (down-sampling ratio of 4/5):
The input frame is then supplied to the optional pre-processing block 102.
Pre-processing block 102 may consist of a high-pass filter with a 50 Hz cut-
off
frequency. High-pass filter 102 removes the unwanted sound components below
50 Hz.
The down-sampled pre-processed signal is denoted by sp(n), n=0,1, 2,
.. ,L-1, where L is the length of the frame (256 at a sampling frequency of
12.8
kHz). In a preferred embodiment of the preemphasis filter 103, the signal
sp(n) is
preemphasized using the following transfer function:
P(z)-1_f~z~

CA 02391562 2005-O1-07
17
where : is a preemphasis factor with a value located between 0 and 1 (a
typical
value is : = 0.7). A higher-order filter could also be used. It should be
pointed
out that high-pass filter 102 and preemphasis filter 103 can be interchanged
to
obtain more efficient fixed-point implementations.
The function of the preemphasis filter 103 is to enhance the high
frequency contents of the input signal. It also reduces the dynamic range of
the
input speech signal, which renders it more suitable for fixed-point
implementation. Without preemphasis, LP analysis in fixed-point using single-
i0 precision arithmetic is difficult to implement.
Preemphasis also plays an important role in achieving a proper overall
perceptual weighting of the quantization error, which contributes to improve
sound quality. This will be explained in more detail herein below.
The output of the preemphasis filter 103 is denoted s(n). This signal is
used for performing LP analysis in calculator module 104. LP analysis is a
technique well known to those .of ordinary skill in the art. In this preferred
embodiment, the autocorrelation approach is used. In the autocorrelation
approach, the signal s(n) is first windowed using a Hamming window (having
usually a leng#h of the order of 30-40 ms). The autocorrelations are computed
from the windowed signal, and Levinson-Durbin recursion is used to compute LP
filter 'coefficients, ai, where i=1,...,p, and where p is the LP order, which
is
typically 16 in wideband coding. The parameters ai are the coefficients of the
transfer function of the LP filter, which is given by the following relation:
P
A(z)=1+~a~ zv
i=1
.
LP analysis is performed in calculator module 104, which also performs
the quantization and interpolation 'of the LP filter coefficients. The LP
filter

CA 02391562 2005-O1-07
18
coefficients are first transformed into another equivalent domain more
suitable for
quantization and interpolation purposes. The line spectral pair (LSP) and
immitance spectral pair (ISP) domains are two domains in which quantization
and
interpolation can be efficiently performed. The 16 LP filter coefficients, ai,
can be
quantized in the order of 30 to 50 bits using split or multi-stage
quantization, or a
combination thereof. The purpose, of the interpolation is to enable updating
the
LP filter coefficients every subframe while transmitting them once every
frame,
which improves the encoder performance without increasing the bit rate.
Quantization and interpolation of the LP filter coefficients is believed to be
otherwise well known to those of ordinary skill in the art and, accordingly,
will not
be further described in the present specification.
The following paragraphs will describe the rest of the coding operations
performed on a subframe ~ basis. In the following description, the filter A(z)
denotes the unquantized interpolated LP filter of the subframe, and the filter
A(z)
denotes the quantized interpolated LP filter of the subframe.
Perceptual Weighting:
In analysis-by-synthesis encoders, the optimum pitch and innovative
parameters are searched by minimizing the mean squared error between the
input speech and synthesized speech in a perceptually weighted domain. This is
equivalent to minimizing the error between the weighted input speech and
weighted synthesis speech.
The weighted signal sw(n) is computed in a perceptual weighting filter
105. Traditionally, the weighted signal sw(n) has been computed by a weighting
filter having a transfer function W(z) in the form:
W(z)=A(zly,)lA(zlyz)
where p < yz < y~ < 1

CA 02391562 2005-O1-07
19
As well known to those of ordinary skill in the art, in prior art analysis-by-
synthesis
(AbS) encoders, analysis shows that the quantization error is weighted by a
transfer function W -1 (z), which is the inverse of the transfer function of
the
perceptual weighting filter 105. This result is well described by B.S. Atal
and M.R.
Schroeder in "Predictive coding of speech and subjective error criteria", IEEE
Transaction ASSP, vol. 27, no. 3, pp. 247-254, June 1979. Transfer function W -
1 (z) exhibits some of the formant structure of the input speech signal. Thus,
the
masking property of the human ear is exploited by shaping the quantization
error
so that it has more energy in the formant regions where it will be masked by
the
strong signal energy present in these regions. The amount of weighting is
controlled by the factors y1 and y2.
The above traditional perceptual weighting filter 105 works well with
telephone band signals. However, it was found that this traditional perceptual
weighting filter 105 is not suitable for efficient perceptual weighting of
wideband
signals. It was also found that the traditional perceptual weighting filter
105 has
inherent limitatioris in modelling the formant structure and the required
spectral
tilt concurrently. The spectral tilt is more pronounced in wideband signals
due to
the wide dynamic range between low and high frequencies. The prior art has
suggested to add. a tilt filter into W(z) in order to control the tilt and
formant
weighting of the virideband input signal separately.
A novel solution to this problem is to introduce the preemphasis filter 103
at the input, compute the LP filter A(z) based on the preempfiasized speech
s(n),
and use a modified filter W(z) by fixing its denominator.

CA 02391562 2005-O1-07
LP analysis is performed in module 104 on the preemphasized signal s(n)
to obtain the LP filter A(z). Also, a new perceptual weighting filter 105 with
fixed
denominator is used. An example of transfer function for the perceptual
weighting
filter 105 is given by the following relation:
5
W(z)-A(z~Y,)~(1-YZZ'~
where p < < < 1
Yz Yr -
10 A higher order can be used at the denominator. This structure substantially
decouples the formant weighting from the tilt.
Note that because A(z) is computed based on the preemphasized speech
signal s(n), the tilt of the filter 1 /A(z/y1 ) is less pronounced compared to
the case
15 when A(z) is computed based on the original speech. Since deemphasis is
performed at the decoder end using a filter having the transfer function:
P-'~Z~ = l~~l -~ z'~~
the quantization error spectrum is shaped by a filter having a transfer
function W
-1 (z)P -1 (z). W hen y2 is set equal to N, which is typically the case, the
spectrum
of the quantization error is shaped by a filter whose transfer function is 1
/A(z/yi ),
with A(z) computed based on the preemphasized speech signal. Subjective
listening showed that this structure for achieving the error shaping by a
combination of preemphasis and modified weighting filtering is very efficient
for
encoding wideband signals, in addition to the advantages of ease of fixed-
point
algorithmic implementation.
Pitch Analysis:

CA 02391562 2005-O1-07
21
In~ order to simplify the pitch analysis, an open-loop pitch lag TOL is first
estimated in the open-loop pitch search module 106 using the weighted speech
signal sw(n). Then the closed-loop pitch analysis, which is performed in
closed-
loop pitch search module 107 on a subframe basis, is restricted around the
open-
loop pitch lag TOL which significantly reduces the search complexity of the
LTP
parameters T and b (pitch lag and pitch gain, respectively). Open-loop pitch
analysis is usually performed in module 106 once every 10 ms (two subframes)
using techniques well known to those of ordinary skill in the art.
The target vector x for LTP (Long Term Prediction) analysis is first
computed. This is usually done by subtracting the zero-input response s0 of
weighted synthesis filter W(z)/A(z) from the weighted speech signal sw(n).
This
zero-input response s0 is calculated by a zero-input response calculator 108.
More specifically, the target vector x is calculated using the following
relation:
x=sw-so
where x is the N-dimensional target vector, sw is the weighted speech vector
in
the subframe, and s0 is the zero-input response of filter W(z)/A(z) which is
the
output of the combined filter W(z)/A(z) due to its initial states. The zero-
input
response calculator 108 is responsive to the quantized interpolated LP filter
A(z)
from the LP analysis, quantization and interpolation calculator module 104 and
to
the initial. states of the weighted synthesis filter W(z)/A(z) stored in
memory
module 111 to calculate the zero-input response s0 (that part of the response
due to the initial states as determined by setting the inputs equal to zero)
of filter
W(z)/A(z). Again, this operation is well known to those of ordinary skill in
the art
and, accordingly, will not be further described.
Of course, alternative but mathematically equivalent approaches can be
used to compute the target vector x.
A N-dimensional impulse response vector h of the weighted synthesis
filter W (z)/A(z) is computed in the impulse response generator module 109
using

CA 02391562 2005-O1-07
22
the LP filter coefficients A(z) and A(z) from module 104. Again, this
operation is
well known to those of ordinary skill in the art and, accordingly, will not be
further
described in the present specification.
The closed-loop pitch (or pitch codebook) parameters b, T and j are
computed in the closed-loop pitch search module 107, which uses the target
vector x, the impulse response vector h and the open-loop pitch lag TOL as
inputs. Traditionally, the pitch prediction has been represented by a pitch
filter
having the following transfer function:
1/(1-bzT)
where b is the pitch gain and T is the pitch delay or lag. In this case, the
pitch
contribution to the excitation signal u(n) is given by bu(n-T), where the
total
excitation is given by
u(n)=bu(n-T)+gck(n)
with g being the innovative codebook gain and ck(n} the innovative codevector
at
index k.
This representation has limitations if the pitch lag T is shorter than the
subframe length N. In another representation, the pitch contribution can be
seen
as a pitch codebook containing the past excitation signal. Generally, each
vector
in the pitch codebook is a shift-by-one version of the previous vector
(discarding
one sample and adding a new sample). For pitch lags T>N, the pitch codebook is
equivalent to the filter structure (1 /(1-bz-T) , and the pitch codebook
vector vT(n)
at pitch lag T is given by
vT (n) = a (n -T) n = 0,..., N-1.

CA 02391562 2005-O1-07
23
For pitch lags T shorter than N, a vector vT(n) is built by repeating the
available
samples from the past excitation until the vector is completed (this is not
equivalent to the filter structure):
In recent encoders, a higher pitch resolution is used which significantly
improves the quality of voiced sound segments. This is achieved by
oversampling the past excitation signal using polyphase interpolation filters.
In
this case, the vector vT(n) usually corresponds to an interpolated version of
the
past excitation, with pitch lag T being a non-integer delay (e.g. 50.25).
The pitch search consists of finding the best pitch lag T and gain b that
minimize the mean squared weighted error E between the target vector x and the
scaled filtered past excitation. Error E being expressed as:
E ° Ilx - bYTll 2
where yT is the filtered pitch codebook vector at pitch lag T:
n
YT ~n~ vT I nl * h~n~ - ~ vT (i)h(n
n = 0,...,N-1.
It can be shown that the error E is minimized by maximizing the search
criterion
C, - x' Yr
Y'T YT

CA 02391562 2005-O1-07
24
where t denotes vector transpose.
In the preferred embodiment of the present invention, a 1/3 subsample
pitch resolution is used, and the pitch (pitch codebook) search is composed of
three stages.
In the first stage, the open-loop pitch lag TOL is estimated in open-loop
pitch search module 106 in response to the weighted speech signal sw(n). As
indicated in the foregoing description, this open-loop pitch analysis is
usually
performed once every 10 ms (two subframes) using techniques well known to
those of ordinary skill in the art.
In the second stage, the search criterion C is searched in the closed-loop
pitch search module 107 for integer pitch lags around the estimated open-loop
pitch lag TOL (usually ~5), which significantly simplifies the search
procedure. A
simple procedure can be used for updating the filtered codevector yT without
the
need to compute the convolution for every pitch lag.
Once an optimum integer pitch lag is found in the second stage, a third
stage of the search (module 107) tests the fractions around that optimum
integer
pitch lag.
When the pitch predictor is represented by a filter of the form 1/(1-bz-T),
which is a valid assumption for pitch lags T>N, the spectrum of the pitch
filter
exhibits a harmonic structure over the entire frequency range, with a harmonic
frequency related to 1/T. In the case of wideband signals, this structure is
not
very efficient since the harmonic structure in wideband signals does not cover
the
entire extended spectrum. The harmonic structure exists only up to a certain
frequency, depending on the speech segment. Thus, in order to achieve
efficient
representation of the pitch contribution in voiced segments of wideband
speech,
the pitch prediction filter needs to have the flexibility of varying the
amount of

i i ii."li ., ,~..I ~i~ L,rnl"N.,~d I.i
CA 02391562 2005-O1-07
periodicity over the wideband spectrum.
A new method which achieves efficient modelling of the harmonic
structure of the speech spectrum of wideband signals is disclosed in the
present
5 specification, whereby several forms of low-pass filters are applied to the
past
excitation and the low-pass filter with higher prediction gain is selected.
When subsample pitch resolution is used, the low-pass filters can be
incorporated into the interpolation filters used to obtain the higher pitch
10 resolution. In this case, the third stage of the pitch search, in which the
fractions
around the chosen integer pitch lag are tested, is repeated for the several
interpolation filters having different low-pass characteristics and the
fraction and
filter index which maximize the search criterion C are selected.
15 A simpler approach is to complete the search in the three stages
described above to determine the optimum fractional pitch lag using only one
interpolation filter with a certain frequency response, and selectthe optimum
low-
pass filter shape at the end by applying the different predetermined low-pass
filters to the chosen pitch codebook vector vT and select the low-pass filter
which
20 minimizes the pitch prediction error. This approach is discussed in detail
below.
Figure 3 illustrates a schematic block diagram of a preferred embodiment
of the proposed approach.
25 In memory module 303, the past excitation signal u(n), n<0, is stored. The
pitch codebook search module 301 is responsive to the target vector x, to the
open-loop pitch lag TOL and to the past excitation signal u(n), n<0, from
memory
module 303 to conduct a pitch codebook (pitch codebook) search minimizing the
above-defined search criterion C. From the result of the search conducted in
module 301, module 302 generates the optimum pitch codebookvectorvT. Note
that since a sub-sample pitch resolution is used (fractional pitch), the past
excitation signal u(n), n<0, is interpolated and the pitch codebook vector vT
corresponds to the interpolated past excitation signal. In this preferred

CA 02391562 2005-O1-07
26
embodiment, the interpolation filter (in module 301, but not shown) has a low-
pass filter characteristic removing the frequency contents above 7000 Hz.
In a preferred embodiment, K filter characteristics are used; these filter
characteristics could be low-pass or band-pass filter characteristics. Once
the
optimum codevector vT is determined and supplied by the pitch codevector
generator 302, K filtered versions of codevector vT are computed respectively
using K different frequency shaping filters such as 305(j), where j = 1, 2,
... , K.
These filtered versions are denoted vf(j) , where j = 1, 2, ... , K. The
different
vectors vf(j) are convolved in respective modules 304(j), where j=0, 1, 2, ...
, K,
with the impulse response h to obtain the vectors yQ), where j=0, 1, 2, ... ,
K. To
calculate the mean squared pitch prediction error for each vector y(j), the
value
y(j) is multiplied by the gain b by means of a corresponding amplifier 307(j)
and
the value byQ) is subtracted from the target vector x by means of a
corresponding
subtractor 308(j). Selector 309 selects the frequency shaping filter 305(j)
which
minimizes the mean squared pitch prediction error
e~) _ Ilx _ b~) yG7~~ Z
j=1, 2,...,K
To calculate the mean squared pitch prediction error e(j) for each value of
y(j),
the value y(j) is multiplied by the gain b by means of a corresponding
amplifier
307(j) and the value b(j)y(j) is subtracted from the target vector x by means
of
subtractors 308(j). Each gain b(j) is calculated in a corresponging gain
calculator
306Q) in association with the frequency shaping filter at index j, using the
following relationship:

CA 02391562 2005-O1-07
27
In selector 309, the parameters b, T, and j are chosen based on vT or vf(j)
which minimizes the mean squared pitch prediction error e.
Referring back to Figure 1, the pitch codebook index T is encoded and
transmitted to multiplexer 112. The pitch gain b is quantized and transmitted
to
multiplexer 112. With this new approach, extra information is needed to encode
the index j of the selected frequency shaping filter in multiplexer 112. For
example, if three filters are used . (j=0, 1, 2, 3), then two bits are needed
to
represent this information. The filter index information j can also be encoded
jointly with the pitch gain b.
Innovative codebook search:
Once the pitch, or LTP (Long Term Prediction) parameters b, T, and j are
determined, the next step is to search for the optimum innovative excitation
by
means of search module 110 of Figure 1. First, the target vector x is updated
by
subtracting the LTP contribution:
x -x-byT
where b is the pitch gain and yT is the filtered pitch codebook vector (the
past
excitation at delay T filtered with the selected low-pass filter and convolved
with
the inpulse response h as described with reference to Figure 3).
The search procedure in CELP is performed by finding the optimum
excitation codevector ck and gain g which minimize the mean-squared-error E
between the target vector and the scaled filtered codevector
E-~~ x -gH~kII2

CA 02391562 2005-O1-07
28
where H is a lower triangular convolution matrix derived from the impulse
response vector h.
In the preferred embodiment of the present invention, the innovative
codebook search is performed in module 110 by means of an algebraic
codebook as described in US patents Nos: 5,444,816 (Adoul et al.) issued on
August 22, 1995; 5,699,482 granted to Adoul et al., on December 17, 1997;
5,754,976 granted to Adoul et al., on May 19, 1998; and 5,701,392 (Adoul et
al.)
dated December 23, 1997.
Once the optimum excitation codevector ck and its gain g are chosen by
module 110, the codebook index k and gain g are encoded and transmitted to
multiplexer 112.
Referring to Figure 1, the parameters b, T, j, A(z), k and g are multiplexed
through the multiplexer 112 before being transmitted through a communication
channel.
Memory update:
In memory module 111 (Figure 1), the states of the weighted synthesis
filter W(z)/A(z) are updated by filtering the excitation signal
a = gck + bvT through the weighted synthesis filter. After this filtering, the
states
of the filter are memorized and used in the next subframe as initial states
for
computing the zero-input response in calculator module 108.
As in the case of the target vector x, other alternative but mathematically
equivalent approaches well known to those of ordinary skill in the art can be
used
to update the filter states.

CA 02391562 2005-O1-07
29
DECODER-200
The speech decoding device 200 of Figure 2 illustrates the various steps
carried out between the digital input 222 (input stream to the demultiplexer
217)
and the output sampled speech 223 (output of the adder 221 ).
Demultiplexer 217 extracts the synthesis model parameters from the
binary information received from a digital input channel. From each received
binary frame, the extracted parameters are:
- the short-term prediction parameters (STP) A(z) (once per frame);
- the long-term prediction (LTP) parameters T, b, and j (for each
subframe); and
- the innovation codebook index k and gain g (for each subframe).
The current speech signal is synthesized based on these parameters as will be
explained hereinbelow.
The innovative codebook 218 is responsive to the index k to produce the
innovation codevector ck, which is scaled by the decoded gain factor g through
an amplifier 224. In the preferred embodiment, an innovative codebook 218 as
described in the above mentioned US patent numbers 5,444,816; 5,699,482;
5,754,976; and 5,701,392 is used to represent the innovative codevector ck .
The generated scaled codevector gck at the output of the amplifier 224 is
processed through a innovation filter 205.
Gain smoothing
At the decoder 200 of Figure 2, a nonlinear gain-smoothing technique is
applied to the innovative codebook gain g in order to improve background noise

CA 02391562 2005-O1-07
performance. Based on the stationarity (or stability) and voicing of the
speech
segment of the wideband signal, the gain g of the innovative codebook 218 is
smoothed in order to reduce fluctuation in the energy of the excitation in
case of
stationary signals. This improves the codec performance in the presence of
5 stationary background noise.
In a preferred embodiment, two parameters are used to control the
amount of smoothing: i.e., the voicing of the subframe of wideband signal and
the
stability of the LP (Linear Prediction) filter 206 both indicative of
stationary
10 background noise in the wideband signal.
Different methods can be used for estimating the degree of voicing in the
subframe.
15 Step 501 (Figure 5):
In a preferred embodiment a voicing factor rv is computed in the voicing
factor generator 204 using the following relation:
20 rv = (Ev - Ec) / (Ev + Ec)
where Ev is the energy of the scaled pitch codevector bvT and Ec is the energy
of the scaled innovative codevector gck. That is
25 N_,
Ev = b2 vT' yr = b2 ~ vT (n)
n=0
and
N-I
~' c - g2 Gkt Gk gl ~ Gk (n)
n=0

CA 02391562 2005-O1-07
31
Note that the value of voicing factor rv lies between -1 and 1, where a value
of 1
corresponds to pure voiced signals and a value of -1 corresponds to pure
unvoiced signals.
Step 502 (Figure 5):
A factor J~ is computed in the gain-smoothing calculator 228 based on rv
through the following relation:
7~= 0.5 ( 1 - rv)
Note that the factor ~, is related to the amount of unvoicing, that is ~,=0
for pure
voiced segments and ~,=1 for pure unvoiced segments.
Step 503 (Figure 5):
A stability factor 6 is computed in a stability factor generator 230 based
on a distance measure which gives the similarity of the adjacent LP filters.
Different similarity measures can be used. In this preferred embodiment, the
LP
coefficients are quantized and interpolated in the Immitance Spectral Pair
(ISP).
It is therefore convenient to derive the distance measure in the ISP domain.
Alternatively, the Line Spectral Frequency (LSF) representation of the LP
filter
can equally be used to find the similarity distance of adjacent LP filters.
Other
measures have also been used in the previous art such as the Itakura measure.
In a preferred embodiment, the ISP distance measure between the ISPs
in the present frame n and the past frame n-1 is calculated in stability
factor
generator 230 and is given by the relation:
p-1
D _ ~,' (ISp ~"~ - isp ~"-~~ ~a
S 1 j
1=

CA 02391562 2005-O1-07
32
where p is the order of the LP filter 206. Note that the first p-1 ISPs being
used
are frequencies in the range 0 to 8p00 Hz.
Step 504 (Figure 5):
The ISP distance measure is mapped in gain-smoothing calculator 228 to
a stability factor A in the range 0 to 1, and derived by
B =1. 25 - DS l 400000. 0
bounded by 0 _< A 51.
Note that larger values of 8 correspond to more stable signals.
Step 505 (Figure 5):
A gain smoothing factor Sm based on both voicing and stability is then
calculated in gain smoothing calculator 228 and is given by
Sm-~B ..
The value of Sm approaches 1 for unvoiced and stable signals, which is the
case
of stationary background noise signals. For pure voiced signals or for
unstable
signals, the value of Sm approaches 0.
Step 506 (Figure 5):
An initial modified gain g0 is computed in gain smoothing calculator 228
by comparing the innovative codebook gain g to a threshold given by the
initial
modified gain from the past subframe, g-1. If g is larger or equal to g-1,
then g0 is
computed by decrementing g by 1.5 dB bounded by g0>_g1. If g is smaller than g-
1, then g0 is computed by incrementing g by 1.5 dB bounded by g0<_g-1. Note

i i N.ili ,~~ ~~ml .I. L~.I,v~n~~i 1 ~ in
CA 02391562 2005-O1-07
33
that incrementing the gain by 1.5 dB is equivalent to multiplying by 1.19. In
other
words
if g<g-1 then ' g0=g*1.19 bounded by g0 __<g-1
and
if g >_g-1 then g0=g/1.19 bounded by g0?g-1
Step 507 (Figure 5):
Finally, the smoothed fixed codebook gain gs is, calculated in gain
smoothing calculator 228 by
gs-Sm*go+~l-Sm~*g
The smoothed gain gs is then used for scaling the innovative
codevector ck in amplifier 232.
Just a word to mention that the above gain smoothing procedure can
be applied to signals other than wideband signals.
Periodicity enhancement:
The generated scaled codevector at the output of the amplifier 224 is
processed through a frequency-dependent pitch enhancer 205.
Enhancing the periodicity of the excitation signal a improves the
quality in case of voiced segments. This was done in the past by filtering the
innovation vector from the innovative codebook (fixed codebook) 218 through a
filter in the form 1/(1-sbz-T) where ~ is a factor below 0.5 which controls
the
amount of introduced periodicity. This approach is less efficient in case of
wideband signals since it introduces periodicity over the entire spectrum. A
new
alternative approach, which is part of the present invention, is disclosed
whereby
periodicity enhancement is achieved by filtering the innovative codevector ck

CA 02391562 2005-O1-07
34
from the innovative (fixed) codebook through an innovation filter 205 (F(z))
whose
frequency response emphasizes the higher frequencies more than lower
frequencies. The coefficients of F(z) are related to the amount of periodicity
in
the excitation signal u.
Many methods known to those skilled in the art are available for
obtaining valid periodicity coefficients. For example, the value of gain b
provides
an indication of periodicity. That is, if gain b is close to 1, the
periodicity of the
excitation signal a is high, and if gain b is less than 0.5, then periodicity
is low.
Another efficient way to derive the filter F(z) coefficients used in a
preferred embodiment, is to relate them to the amount of pitch contribution in
the
total excitation signal u. This results in a frequency response depending on
the
subframe periodicity, where higher frequencies are more strongly emphasized v
(stronger overall slope) for higher pitch gains. Innovation filter 205 has the
effect
of lowering the energy of the innovative codevector ck at low frequencies when
the excitation signal a is more periodic, which enhances the periodicity of
the
excitation signal a at lower frequencies more than higher frequencies.
Suggested forms for innovation filter 205 are
(1) F(z)=I-6Z~, (2) F(z)=-c~z+1-az'
or
where a or a are periodicity factors derived from the level of periodicity of
the
excitation signal u.
The second three-terrii form of F(z) is used in a preferred
embodiment. The periodicity factor a is computed in the voicing factor
generator
204. Several methods can be used to derive the periodicity factor a based on
the periodicity of the excitation signal u. Two methods are presented below.
Method 1:

CA 02391562 2005-O1-07
The ratio of pitch contribution to the total excitation signal a is first
computed in voicing factor generator 204 by
5
N-!
a r bl ~ vT ~n)
b VT VT n=0
RP = ut a - N_I
~u~ (~~
n=0
where vT is the pitch codebook vector, b is the pitch gain, and a is the
excitation
signal a given at the output of the adder 219 by
a = gck + bvT
Note that the term bvT has its source in the pitch codebook (adaptive
codebook) 201 in response to the pitch lag T and the past value of a stored in
memory 203. The pitch codevector vT from the pitch codebook 201 is then
processed through a low-pass filter 202 whose cut-off frequency is adjusted by
means of the index j from the demultiplexer 217. The resulting codevector vT
is
then multiplied by the gain b from the demultiplexer 217 through an amplifier
226
to obtain the signal bvT.
The factor a is calculated in voicing factor generator 204 by
a = qRp bounded by a < q
where q is a factor which controls the amount of enhancement (q is set to 0.25
in
this preferred embodiment).

,., i ", ., , ~. ,
CA 02391562 2005-O1-07
36
Method 2:
Another method used in a preferred embodiment of the invention for
calculating periodicity factor a is discussed below.
First, a voicing factor rv is computed in voicing factor generator 204 by
r~ _ (Ev - E~~ ~ (Ev + E~~
where Ev is the energy of the scaled pitch codevector bvT and Ec is the energy
of the scaled innovative codevector gck. That is
N-1
Ev = b~ vrt yr = b2 ~ vT Vin)
n=0
and
N-1
Ec - gZ Ckt Ck - g2 ~ Ck ~~)
n=0
, Note that the value of rv lies between -1 and 1 (1 corresponds to
purely voiced signals and -1 corresponds to purely unvoiced signals).
In this preferred embodiment, the factor a is then computed in voicing
factor generator 204 by
a = 0.125 (1 + nr)

CA 02391562 2005-O1-07
37
which corresponds to a value of 0 for purely unvoiced signals and 0.25 for
purely
voiced signals.
In the first, two-term form of F(z), the periodicity factor Q can be
approximated by using 6= 2a in methods .1 and 2 above. In such a case, the
periodicity factor a is calculated as follows in method 1 above:
a = 2qRp bounded by a < 2q.
In method 2, the periodicity factor a is calculated as follows:
a = 0.25 (1 + rv).
The enhanced signal cf is therefore computed by filtering the scaled
innovative codevector gck through the innovation filter 205 (F(z)).
The enhanced excitation signal u' is computed by the adder 220 as:
u'=cf+bvT
_
Note that this process is not performed at the encoder 100: Thus, it is
essential to update the content of the pitch codebook 201 using the excitation
signal a without enhancement to keep synchronism between the encoder 100
and decoder 200. Therefore, the excitation signal a is used to update the
memory 203 of the pitch codebook 201 and the enhanced excitation signal u' is
used at the input of the LP synthesis filter 206.

CA 02391562 2005-O1-07
38
Synthesis and deemphasis
The synthesized signal s' is computed by filtering the enhanced
excitation signal u' through the LP synthesis filter 206 which has the form 1
/A(z),
where A(z) is the interpolated LP filter in the current subframe. As can be
seen in
Figure 2, the quantized LP coefficients A(z) on line 225 from demultiplexer
217
are supplied to the LP synthesis filter 206 to adjust the parameters of the LP
synthesis filter 206 accordingly. The deemphasis filter 207 is the inverse of
the
preemphasis filter 103 of Figure 1. The transferfunction of the deemphasis
filter
207 is given by
D(z)=1 ~~l-fcz'~~
where : is a preemphasis factor with a value located between 0 and 1 (a
typical
value is : = 0.7). A higher-order filter could also be used.
The vector s' is filtered through the deemphasis filter D(z) (module
207) to obtain the vector sd, which is passed through the high-pass filter 208
to
remove the unwanted frequencies below 50 Hz and further obtain sh.
Oversampling and high-frequency regeneration
The over-sampling module 209 conducts the inverse process of the
down-sampling module 101 of Figure 1. In this preferred embodiment,
oversampling converts from the 12.8 kHz sampling rate to the original 16 kHz
sampling rate, using techniques well known to those of ordinary skill in the
art.
The oversampled synthesis signal is denoted ~. Signal S is also referred to as
the synthesized wideband intermediate signal. '
The oversampled synthesis ~ signal does not contain the higher
frequency components which were lost by the downsampling process (module

CA 02391562 2005-O1-07
39
101 of Figure 1 ) at the encoder 100. This gives a low-pass perception to the
synthesized speech signal. To restore the full band of the original signal, a
high
frequency generation procedure is disclosed. This procedure is performed in
modules 210 to 216, and adder 221, and requires input from voicing factor
generator 204 (Figure 2).
In this new approach, the high frequency contents are generated by
filling the upper part of the spectrum with a white noise properly scaled in
the
excitation domain, then converted to the speech domain, preferably by shaping
it
with the same LP synthesis filter used for synthesizing the down-sampled
signal
The high frequency generation procedure is described hereinbelow
The random noise generator 213 generates a white noise sequence
w' with a flat spectrum over the entire frequency bandwidth, using techniques
well known to those of ordinary skill~in the art. The generated sequence has a
length N' which is the subframe length in the original domain. Note that N is
the
subframe length in the down-sampled domain. In this preferred embodiment,
N=64 and N'=80 which correspond to 5 ms.
The white noise sequehce is properly scaled in the gain adjusting
module 214. Gain adjustment comprises the following steps. First, the energy
of
the generated noise sequence w' is set equal to the energy of the enhanced
excitation signal u' computed by an energy computing module 210, and the
resulting scaled noise sequence is given by
N_~
u.a (n)
w(n) = w'(n)
~ H,.a(~)
' n=0
n=0,...,N'-1.

","
CA 02391562 2005-O1-07
The second step in the gain scaling is to take into account the high
frequency contents of the synthesized signal at the output of the voicing
factor
generator 204 so as to reduce the energy of the generated noise in case of
5 voiced segments (where less energy is present at high frequencies compared
to -
unvoiced segments). In this preferred embodiment, measuring the high
frequency contents is implemented by measuring the tilt of the synthesis
signal
through a spectral tilt calculator 212 and reducing the energy accordingly.
Other
measurements such as zero crossing measurements can equally be used.
10 When the tilt is very strong, which corresponds to voiced segments, the
noise
energy is further reduced. The tilt factor is computed in module 212 as the
first
correlation coefficient of the synthesis signal sh and it is given by:
N-!
~'h (n)sh (~ -1)
tilt = °°I
N-1 2
E Sh ~Yl~
n=0
conditioned by tilt ?O and tilt >_ rv.
where voicing factor rv is given by
rv = (E~ - E~) ~ (Ev + E~)
where Ev is the energy of the scaled pitch codevector bvT and Ec is the energy
of the scaled innovative codevector gck, as described earlier. Voicing factor
nr is
most often less than tilt but this condition was introduced as a precaution
against
high frequency tones where the tilt value is negative and the value of rv is
high.
Therefore, this condition reduces the noise energy for such tonal signals.

. .. .", . ~ J.~~~.n l,a.,. .. .n
CA 02391562 2005-O1-07
The tilt value is 0 in case of flatspectrum and 1 in case of strongly voiced
signals, and it is negative in case' of unvoiced signals where more energy is
present at high frequencies.
Different methods can be used to derive the scaling factor gt from the
amount of high frequency contents. In this invention, two methods are given
based on the tilt of signal described above.
Method 1:
The scaling factor gt is derived from the tilt by
gt = 1 - tilt bounded by 0.2 S gt <_i .0
For strongly voiced signal where the tilt approaches 1, gt is 0.2 and for
strongly
unvoiced signals gt becomes 1Ø
Method 2:
The tilt factor gt is first restricted to be larger or equal to zero, then the
scaling factor is derived from the tilt by
gr - l0-o.srtrr
The scaled noise sequence wg produced in gain adjusting module 214 is
therefore given by:
wg = gt w.
W hen the tilt is close to zero, the scaling factor gt is close to 1, which

"~." i, "~ .~ ,n, w , "~
CA 02391562 2005-O1-07
42
does not result in energy reduction. When the tilt value is 1, the scaling
factor gt
results in a reduction of 12 dB in the energy of the generated noise.
Once the noise is properly scaled (wg ), it is brought into the speech
domain using the spectral shaper 215. In the preferred embodiment, this is
achieved by filtering the noise wg through a bandwidth expanded version of the
same LP synthesis filter used in the down-sampled domain (1/A(z/0.8)). The
corresponding bandwidth expanded LP filter coefficients are calculated in
spectral shaper 215.
The filtered scaled noise sequence wf is then band-pass filtered to the
required frequency range to be restored using the band-pass filter 216. In the
preferred embodiment, the band-pass filter 216 restricts the noise sequence to
the frequency range 5.6-7.2 kHz. The resulting band-pass filtered noise
sequence z is added in adder 221 to the oversampled synthesized speech signal
s' to obtain the final reconstructed sound signal. sout on the output 223.
Although the present invention has been described hereinabove by way
of a preferred embodiment thereof, this embodiment can be modified at will,
within the scope of the appended claims, without departing from the spirit and
nature of the subject inveqtion. Even though the preferred embodiment
discusses the use of wideband speech signals, it will be obvious to those
skilled
in the art that the subject invention is also directed to other embodiments
using
wideband signals in general and that it is not necessarily limited to speech
applications.

Dessin représentatif