Selection of scalar quantixation (SQ) and vector quantization (VQ) for speech coding

• US 8,620,647 B2
• Filed: 01/26/2009
• Issued: 12/31/2013
• Est. Priority Date: 09/18/1998
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: WIAV Solutions LLC
• Inventors: Gao, Yang, Benyassine, Adil
• Primary Examiner(s): Opsasnick, Michael N
• Application Number:
  US12/321,935
• Publication Number:
  US 20090182558A1
• Time in Patent Office: 1,800 Days
• Field of Search: 704214-219
• US Class Current: 704/214
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/09 Long term prediction, i.e. removing periodical redundancies, e.g. by using adaptive codebook or pitch predictor
• CPC Class Codes: G10L 19/12 the excitation function being a code excitation, e.g. in code excited linear prediction [CELP] vocoders
• CPC Class Codes: G10L 19/18 Vocoders using multiple modes
• CPC Class Codes: G10L 19/20 using sound class specific coding, hybrid encoders or object based coding
• CPC Class Codes: G10L 2019/0002 Codebook adaptations
• CPC Class Codes: G10L 2019/0016 Codebook for LPC parameters
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals

Abstract

In accordance with one aspect of the invention, a selector supports the selection of a first encoding scheme or the second encoding scheme based upon the detection or absence of the triggering characteristic in the interval of the input speech signal. The first encoding scheme has a pitch pre-processing procedure for processing the input speech signal to form a revised speech signal biased toward an ideal voiced and stationary characteristic. The pre-processing procedure allows the encoder to fully capture the benefits of a bandwidth-efficient, long-term predictive procedure for a greater amount of speech components of an input speech signal than would otherwise be possible. In accordance with another aspect of the invention, the second encoding scheme entails a long-term prediction mode for encoding the pitch on a sub-frame by sub-frame basis. The long-term prediction mode is tailored to where the generally periodic component of the speech is generally not stationary or less than completely periodic and requires greater frequency of updates from the adaptive codebook to achieve a desired perceptual quality of the reproduced speech under a long-term predictive procedure.

Claims

1. A method of coding a speech signal using a multi-rate speech coder having an adaptive codebook, a fixed codebook, and a coding rate selected from a plurality of coding rates including a first coding rate and a second coding rate, the method comprising:

obtaining an adaptive codebook gain;

obtaining a fixed codebook gain;

scalar quantizing the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the coding rate is the first coding rate, to generate a first quantized adaptive codebook gain and a first quantized fixed codebook gain;

vector quantizing the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the Coding rate is the second coding rate, to generate a second quantized adaptive codebook gain and a second quantized fixed codebook gain;

converting the speech signal into a first encoded speech using the first quantized adaptive codebook gain and the first quantized fixed codebook gain if the coding rate is the first coding rate; and

converting the speech signal into a second encoded speech using the second quantized adaptive codebook gain and the second quantized fixed codebook gain if the coding rate is the second coding rate.

2. The method of claim 1, wherein the first coding rate is higher than the second coding rate.

3. The method of claim 1, wherein the vector quantizing further comprises predicting the fixed codebook gain.

4. The method of claim 1, wherein the vector quantizing further comprises minimizing a mean squared error between the speech signal and a reconstructed speech signal if the coding rate is the second coding rate.

5. The method of claim 1, wherein the scalar quantizing further comprises using four (4) bits for the first quantized adaptive codebook gain and five (5) bits for the first quantized fixed codebook gain if the coding rate is the first coding rate.

6. The method of claim 1, wherein the vector quantizing further comprises using seven (7) bits for the second quantized adaptive codebook gain and the second quantized fixed codebook gain if the coding rate is the second coding rate.

7. The method of claim 6, wherein the plurality of coding rates further includes a third coding rate, and wherein the method further comprises:

vector quantizing the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the coding rate is the third coding rate, to generate a third quantized adaptive codebook gain and a third quantized fixed codebook gain;

converting the speech signal into a third encoded speech using the third quantized fixed codebook gain and the third quantized adaptive codebook gain if the coding rate is the third coding rate;

wherein the vector quantizing uses six (6) bits for the third quantized adaptive codebook gain and the third quantized fixed codebook gain, if the coding rate is the third coding rate.

8. The method of claim 7, wherein the second coding rate is higher than the third coding rate.

9. A multi-rate speech coder for coding a speech signal, the multi-rate speech coder using a coding rate selected from a plurality of coding rates including a first coding rate and a second coding, the multi-rate speech coder comprising:

an adaptive codebook;

a fixed codebook;

a speech processing circuitry configured to;

obtain an adaptive codebook gain;

obtain a fixed codebook gain;

scalar quantize the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the coding rate is the first coding rate, to generate a first quantized adaptive codebook gain and a first quantized fixed codebook gain;

vector quantize the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the coding rate is the second coding rate, to generate a second quantized adaptive codebook gain and a second quantized fixed codebook gain;

convert the speech signal into a first encoded speech using the first quantized adaptive codebook gain and the first quantized fixed codebook gain if the coding rate is the first coding rate; and

convert the speech signal into a second encoded speech using the second quantized adaptive codebook gain and the second quantized fixed codebook gain if the coding rate is the second coding rate.

10. The multi-rate speech coder of claim 9, wherein the first coding rate is higher than the second coding rate.

11. The multi-rate speech coder of claim 10, wherein speech processing circuitry is further configured to predict the fixed codebook gain.

12. The multi-rate speech coder of claim 9, wherein the speech processing circuitry is configured to vector quantize by minimizing a mean squared error between the speech signal and a reconstructed speech signal if the coding rate is the second coding rate.

13. The multi-rate speech coder of claim 9, wherein the speech processing circuitry uses four (4) bits for the first adaptive codebook gain and five (5) bits for the first fixed codebook gain if the coding rate is the first coding rate.

14. The multi-rate codec of claim 9, wherein the speech processing circuitry uses seven (7) bits for the second adaptive codebook gain and the second fixed codebook gain if the coding rate is the second coding rate.

15. The multi-rate speech coder of claim 9, wherein the plurality of coding rates further includes a third coding rate, and wherein the speech processing circuitry is further configured to:

vector quantize the adaptive codebook gain and the fixed codebook gain for coding the speech signal, if the coding rate is the third coding rate, to generate a third quantized adaptive codebook gain and a third quantized fixed codebook gain;

convert the speech signal into a third encoded speech using the third quantized fixed codebook gain and the third quantized adaptive codebook gain if the coding rate is the third coding rate;

wherein the vector quantizing uses six (6) bits for the third quantized adaptive codebook gain and the third quantized fixed codebook gain, if the coding rate is the third coding rate.

16. The multi-rate speech coder of claim 15, wherein the second coding rate is higher than the third coding rate.