Frequency band extension in an audio signal decoder
The invention relates to a method for extending the frequency band of an audio signal during a decoding or improvement process comprising a step of decoding or extracting, in a first so-called low frequency band, an excitation signal and coefficients of a linear prediction filter. The method comprises the following steps: —obtaining a signal extended in at least a second frequency band higher than the first frequency band from an oversampled excitation signal extended in at least a second frequency band; —scaling the extended signal by means of a gain defined by subframe on the basis of an energy ratio of a frame and of a subframe; —filtering said scaled extended signal with a linear prediction filter of which the coefficients are derived from the coefficients of the low frequency band filter. The invention also relates to a frequency band extension device implementing the described method and a decoder comprising such a device.
1. A frequency band extending method applied to an audio frequency signal in a decoding process comprising an act of decoding or an enhancement process comprising an act of extraction, in a first frequency band called low band, of an excitation signal and coefficients of a linear prediction filter, wherein the method comprises:
obtaining of an extended signal from an oversampled and extended excitation signal in at least one second frequency band higher than the first frequency band generated according to the following equation:
U
HB
1
(
k
)
=
{
0
k
=
0
,
⋯
,
199
U
(
k
)
k
=
200
,
⋯
,
239
U
(
k
+
start_band
-
240
)
k
=
240
,
⋯
,
319
with k being an index of a signal sample, U HB1 (k) being a spectrum of the extended excitation signal, U(k) being a spectrum of the excitation signal obtained after a time-frequency transform act and start_band being a predefined variable;
scaling of the extended signal by a gain defined per sub-frame using a calculation based on a comparison between the extended signal and the low-band signal of a ratio of energy per sub-frame and energy per frame so that the extended signal has the same ratio of energy between a sub-frame and a frame as in the low-band signal;
filtering of said scaled extended signal by a linear prediction filter whose coefficients are derived from decoded or extracted coefficients of a linear prediction filter in the low-band.
2. The method as claimed in claim 1 , wherein the method further comprises an adaptive bandpass filtering as a function of decoding bit rate of a current frame.
3. The method as claimed in claim 2 , further comprising a de-emphasis filtering of the extended signal at least in the second frequency band.
4. The method as claimed in claim 1 , wherein the method comprises a time-frequency transform of a time excitation signal, the act of obtaining of an extended signal then being performed in the frequency domain and an inverse time-frequency transform of the obtained extended signal before the scaling and filtering steps.
5. The method as claimed in claim 4 , further comprising a de-emphasis filtering of the extended signal at least in the second frequency band.
6. The method as claimed in claim 1 , wherein the method comprises a de-emphasis filtering of the extended signal at least in the second frequency band.
7. The method as claimed in claim 1 , wherein the method further comprises a generation of a noise signal at least in the second frequency band, the extended signal being obtained by combination of the extended excitation signal and of the noise signal.
8. The method as claimed in claim 7 , wherein the combination is performed by adaptive additive mixing with a level equalization gain between the extended excitation signal and the noise signal.
9. A frequency band extending device for extending the frequency band of an audio frequency signal comprising a decoding module for decoding or an extraction module for extracting, in a first frequency band called low band, an excitation signal and coefficients of a linear prediction filter, wherein the device comprises:
a module for obtaining an extended signal from an oversampled and extended excitation signal in at least one second frequency band higher than the first frequency band generated according to the following equation:
U
HB
1
(
k
)
=
{
0
k
=
0
,
⋯
,
199
U
(
k
)
k
=
200
,
⋯
,
239
U
(
k
+
start_band
-
240
)
k
=
240
,
⋯
,
319
with k being an index of a signal sample, U HB1 (k) being a spectrum of the extended excitation signal, U(k) being a spectrum of the excitation signal obtained after a time-frequency transform act and start_band being a predefined variable;
a module for scaling the extended signal by a gain defined per sub-frame using a calculation based on a comparison between the extended signal and the low-band signal of a ratio of energy per sub-frame and energy per frame so that the extended signal has the same ratio of energy between a sub-frame and a frame as in the low-band signal;
a module for filtering said scaled extended signal by a linear prediction filter whose coefficients are derived from decoded or extracted coefficients of a linear prediction filter in the low-band.
10. An audio frequency signal decoder, comprising a frequency band extension device for extending the frequency band of an audio frequency signal comprising a decoding module for decoding or an extraction module for extracting, in a first frequency band called low band, an excitation signal and coefficients of a linear prediction filter, wherein the device comprises:
a module for obtaining an extended signal from an oversampled and extended excitation signal in at least one second frequency band higher than the first frequency band generated according to the following equation:
U
HB
1
(
k
)
=
{
0
k
=
0
,
⋯
,
199
U
(
k
)
k
=
200
,
⋯
,
239
U
(
k
+
start_band
-
240
)
k
=
240
,
⋯
,
319
with k being an index of a signal sample, U HB1 (k) being a spectrum of the extended excitation signal, U(k) being a spectrum of the excitation signal obtained after a time-frequency transform act and start_band being a predefined variable;
a module for scaling the extended signal by a gain defined per sub-frame using a calculation based on a comparison between the extended signal and the low-band signal of a ratio of energy per sub-frame and energy per frame so that the extended signal has the same ratio of energy between a sub-frame and a frame as in the low-band signal;
a module for filtering said scaled extended signal by a linear prediction filter whose coefficients are derived from decoded or extracted coefficients of a linear prediction filter in the low-band.
11. A non-transitory storage medium that can be read by a frequency band extending device for extending the frequency band of an audio frequency signal on which is stored a computer program comprising code instructions for execution of a frequency band extending method, the method for extending the frequency band applied to an audio frequency signal in a decoding process comprising an act of decoding or an enhancement process comprising an act of extraction, in a first frequency band called low band, of an excitation signal and coefficients of a linear prediction filter, wherein the method comprises:
obtaining of an extended signal from an oversampled and extended excitation signal in the at least one second frequency band higher than the first frequency band generated according to the following equation:
U
HB
1
(
k
)
=
{
0
k
=
0
,
⋯
,
199
U
(
k
)
k
=
200
,
⋯
,
239
U
(
k
+
start_band
-
240
)
k
=
240
,
⋯
,
319
with k being an index of a signal sample, U HB1 (k) being a spectrum of the extended excitation signal, U(k) being a spectrum of the excitation signal obtained after a time-frequency transform act and start_band being a predefined variable;
scaling of the extended signal by a gain defined per sub-frame using a calculation based on a comparison between the extended signal and the low-band signal of a ratio of energy per sub-frame and energy per frame so that the extended signal has the same ratio of energy between a sub-frame and a frame as in the low-band signal;
filtering of said scaled extended signal by a linear prediction filter whose coefficients are derived from decoded or extracted coefficients of a linear prediction filter in the low-band.