Efficient temporal envelope coding approach by prediction between low band signal and high band signal

• US 8,352,279 B2
• Filed: 09/04/2009
• Issued: 01/08/2013
• Est. Priority Date: 09/06/2008
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: Gao, Yang
• Primary Examiner(s): He, Jialong
• Application Number:
  US12/554,868
• Publication Number:
  US 20100063812A1
• Time in Patent Office: 1,222 Days
• Field of Search: 704500-504, 704/219, 704/225, 704/230
• US Class Current: 704/500
• CPC Class Codes: G10L 19/002 Dynamic bit allocation for perceptual audio coders G10L19/032
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/025 Detection of transients or attacks for time/frequency resolution switching
• CPC Class Codes: G10L 19/04 using predictive techniques

Abstract

This invention proposes a more efficient way to quantize temporal envelope shaping of high band signal by benefiting from energy relationship between low band signal and high band signal; if low band signal is well coded or it is coded with time domain codec such as CELP, temporal envelope shaping information of low band signal can be used to predict temporal envelope shaping of high band signal; the temporal envelope shaping prediction can bring significant saving of bits to precisely quantize temporal envelope shaping of high band signal. This prediction approach can be combined with other specific approach to further increase the efficiency and save mores bits.

Claims

1. An audio/speech signal encoding and decoding method, comprising:

predicting a temporal energy envelope shaping of a high frequency band signal from a low frequency band signal;

estimating an energy ratio between the high frequency band signal and the low frequency band signal, and quantizing the energy ratio;

sending the low frequency band signal and the quantized energy ratio from an encoder to a decoder;

receiving the low frequency band signal and the quantized energy ratio at the decoder;

estimating a temporal energy envelope shaping of the low frequency band signal from the received low frequency band signal;

decoding the quantized energy ratio between the high frequency band signal and the low frequency band signal;

multiplying the temporal energy envelope shaping of the low frequency band signal with the quantized energy ratio between the high frequency band signal and the low frequency to obtain a predicted temporal energy envelope shaping of the high frequency band signal; and

obtaining the high frequency band signal according to the predicted temporal energy envelope shaping of the high frequency band signal.

2. The method according to claim 1, further comprising:

obtaining the high frequency band signal and the low frequency band signal by splitting an input signal.

3. The method according to claim 1, wherein the low frequency band signal has a plurality of frames, and each of the plurality of frames has a plurality of sub-segments;

and wherein predicting the temporal energy envelope shaping of the high frequency band signal from the low frequency band signal comprises;

calculating a square root of an average energy of each sub-segment in a Linear domain or a Log domain, to obtain a plurality of energy magnitudes; and

applying the plurality of energy magnitudes to form the temporal energy envelope shaping of the high frequency band signal.

4. The method according to claim 3, wherein a duration of each sub-segment is 1.25 milliseconds.

5. The method according to claim 1, wherein the high frequency band signal and the low frequency band signal respectively have a plurality of frames, and each of the plurality of frames has a plurality of sub-segments;

and wherein the energy ratio between the high frequency band signal and the low frequency band signal is estimated at least once per frame.

6. The method according to claim 5, wherein energy ratios for the sub-segments between a current frame and a previous frame are interpolated in a Log domain or a Linear domain.

7. An audio/speech signal encoding method, comprising:

predicting a temporal energy envelope shaping of a high frequency band signal from a low frequency band signal;

estimating an energy ratio between the high frequency band signal and the low frequency band signal, and quantizing the energy ratio;

sending the low frequency band signal and the quantized energy ratio from an encoder to a decoder;

multiplying a temporal energy envelope shaping of the low frequency band signal with the quantized energy ratio to obtain the predicted temporal energy envelope shaping of the high frequency band signal;

estimating correction errors of the predicted temporal energy envelope shaping of the high frequency band signal by comparing it with an ideal temporal energy envelope shaping of the high frequency band signal at the encoder;

quantizing the correction errors; and

sending the quantized correction errors to the decoder.

8. An audio/speech signal decoding method, comprising:

receiving a low frequency band signal from an encoder;

estimating a temporal energy envelope shaping of the low frequency band signal from the received low frequency band signal;

receiving an energy ratio between a high frequency band signal and the low frequency band signal;

multiplying the temporal energy envelope shaping of the low frequency band signal with the received energy ratio to obtain a predicted temporal energy envelope shaping of the high frequency band signal;

obtaining the high frequency band signal according to the predicted temporal energy envelope shaping of the high frequency band signal;

estimating correction errors of the predicted temporal energy envelope shaping of the high frequency band signal according to received information from the encoder; and

obtaining the high frequency band signal according to the predicted and corrected temporal energy envelope shaping of the high frequency band signal.

9. A codec, comprising an encoder and a decoder;

wherein the encoder comprises a processor and a transmitter,

the processor is configured to;

predict at least one temporal energy envelope shaping of at least one high frequency band signal from at least one low frequency band signal; and

estimate at least one energy ratio between the at least one high frequency band signal and the at least on low frequency band signal, and quantizing the at least one energy ratio;

and the transmitter is configured to;

transmit the at least one low frequency band signal and the at least one quantized energy ratio to a decoder;

wherein the decoder comprises a receiver and a processor,

the receiver is configured to receive the low frequency band signal and the quantized energy ratio;

and the processor is configured to;

estimate a temporal energy envelope shaping of the low frequency band signal from the received low frequency band signal;

decode the quantized energy ratio between the high frequency band signal and the low frequency band signal;

multiply the temporal energy envelope shaping of the low frequency band signal with the quantized energy ratio between the high frequency band signal and the low frequency to obtain a predicted temporal energy envelope shaping of the high frequency band signal; and

obtain the high frequency band signal according to the predicted temporal energy envelope shaping of the high frequency band signal.