Gain quantization for a CELP speech coder

Details Gain quantization for a CELP speech coder Gain quantization for a CELP speech coder

2000-05-19

2004-08-24

To, Doris H. (Department: 2654)

Data processing: speech signal processing, linguistics, language

Speech signal processing

For storage or transmission

C704S219000

Reexamination Certificate

active

06782360

ABSTRACT:

BACKGROUND OF THE INVENTION
The field of the present invention relates generally to the coding of speech in voice communication systems and, more particularly to an improved code-excited linear prediction coding system and method of coding the gain quantization parameters of a speech signal with fewer bits.
To model basic speech sounds, speech signals are sampled over time and stored in frames as a discrete waveform to be digitally processed. However, in order to increase the efficient use of the communication bandwidth for speech, speech is coded before being transmitted especially when speech is intended to be transmitted under limited bandwidth constraints. Numerous algorithms have been proposed for the various aspects of speech coding. In coding speech, the speech coding algorithm tries to represent characteristics of the speech signal in a manner which requires less bandwidth. For example, the speech coding algorithm seeks to remove redundancies in the speech signal. A first step is to remove short-term correlations. One type of signal-coding technique is linear predictive coding (LPC). In using a LPC approach, the speech signal value at any particular time is modeled as a linear function of previous values. By using a LPC approach, short-term correlations can be reduced and efficient speech signal representations can be determined by estimating and applying certain prediction parameters to represent the signal. After the removal of short-term correlations in a speech signal, a LPC residual signal remains. This residual signal contains periodicity information that needs to be modeled. The second step in removing redundancies in speech is to model the periodicity information. Periodicity information may be modeled by using pitch prediction. Certain portions of speech have periodicity while other portions do not. For example, the sound “aah” has periodicity information while the sound “shhh” has no periodicity information.
In applying the LPC technique, a conventional source encoder operates on speech signals to extract modeling and parameter information to be coded for communication to a conventional source decoder via a communication channel. One way to code modeling and parameter information into a smaller amount of information is to use quantization. Quantization of a parameter involves selecting the closest entry in a table or codebook to represent the parameter. Thus, for example, a parameter of 0.125 may be represented by 0.1 if the codebook contains 0, 0.1, 0.2, 0.3, etc. Quantization includes scalar quantization and vector quantization. In scalar quantization, one selects the entry in the table or codebook that is the closest approximation to the parameter, as described above. By contrast, vector quantization combines two or more parameters and selects the entry in the table or codebook which is closest to the combined parameters. For example, vector quantization may select the entry in the codebook that is the closest to the difference between the parameters. A codebook used to vector quantize two parameters at once is often referred to as a two-dimensional codebook. A n-dimensional codebook quantizes n parameters at once.
In CELP (Code Excited Linear Prediction) speech coding, there are two types of gain. The first type of gain is the pitch gain G
P
, also known as the adaptive codebook gain. The adaptive codebook gain is sometimes referred to, including herein, with the subscript “a” instead of the subscript “p”. The second type of gain is the fixed codebook gain G
C
. Speech coding algorithms have quantized parameters including the adaptive codebook gain and the fixed codebook gain. Once coded, the parameters representing the input speech signal are transmitted to a transceiver.
At the transceiver, a decoder receives the coded information. Because the decoder is configured to know the manner in which speech signals are encoded, the decoder decodes the coded information to reconstruct a signal for playback that sounds to the human ear like the original speech.
Therefore, transmitting the coded modeling and parameter information to the decoder requires a certain amount of valuable communication channel bandwidth. In order to increase the efficient use of the bandwidth, improvements to the manner in which modeling and parameter information is coded are needed. Coding algorithms need to reduce the amount of information in bits that must be transmitted over the communication channel. However, there is a countervailing need for a coding algorithm that not only reduces the amount of information in bits that must be communicated over the channel, but also maintains a high quality level of the reproduced speech.
SUMMARY OF THE INVENTION
Various separate aspects of the present invention can be found in a speech encoding system and method that uses an analysis-by-synthesis coding approach on a speech signal. The speech encoding system has an encoder processor and a plurality of codebooks that generate excitation vectors. The speech encoder analyzes and classifies each frame of speech into periodic-like speech or non-periodic like speech. For simplicity throughout this application and claims, periodic-like signals and periodic signals are referred to as “periodic” signals while non-periodic speech is referred to as “non-periodic” or “not periodic” signals.
There are at least three main alternative embodiments as described below. A first embodiment uses a new gain quantization strategy for periodic speech and uses a known gain quantization approach for non-periodic speech. The second embodiment uses the new gain quantization strategy for both periodic speech and non-periodic speech where the bit rate (number of bits per second) for non-periodic speech is greater than that for periodic speech, but less than the bit rate resulting from known gain quantization approaches. The third embodiment uses the new gain quantization strategy for all speech which results in a bit rate equivalent to that for non-periodic speech in the second embodiment.
The first embodiment is described first below, followed by the second and third embodiments. If the speech is periodic, the pitch gains are derived from the original unquantized weighted speech signal before closed loop subframe processing begins. This is different from the traditional way where the pitch gains are derived from the closed loop subframe processing. A “closed loop” process finds the vector in a codebook that generates synthesized speech that is closest perceptually to the original input speech. By contrast, an “open loop” process finds the vector in a codebook that is closest to the gain vector (or a transformed gain vector such as the log of the gain vector). In an open loop process, the closeness of two vectors does not depend on how perceptually close the synthesized speech is to the original speech. The speech encoder performs a different gain quantization process depending if the speech is periodic or not. If the speech is periodic, the improved speech encoder performs the following two gain quantizations: (1) perform a pre-vector quantization of the adaptive codebook gain G
P
for each subframe of the frame which is based on the original unquantized weighted speech; this quantization occurs before the closed loop subframe processing begins; and (2) perform a closed-loop delayed decision vector quantization of the fixed codebook gain G
C
at the end of the subframe processing.
A first, separate aspect of the present invention is a speech encoder that classifies speech into periodic-like and non-periodic like speech and processes gain quantization of periodic-like speech differently than that of non-periodic like speech.
A second, separate aspect of the present invention is a speech encoder that performs for each frame of periodic speech a pre-vector quantization of the G
P
for each subframe of the frame and performs a closed-loop delayed decision vector quantization of the G
C
.
A third, separate aspect of the present invention is a speech encoder that performs a closed loop delayed decision vector quantization of the

Affiliated with

Also associated with

Rating

Gain quantization for a CELP speech coder does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Gain quantization for a CELP speech coder, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Gain quantization for a CELP speech coder will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3347750