Method and apparatus for predictively quantizing voiced speech with substraction of weighted parameters of previous frames

• US 20040260542A1
• Filed: 07/22/2004
• Published: 12/23/2004
• Est. Priority Date: 04/24/2000
• Status: Active Grant

• Current Assignee: Andrew P. Dejaco, Arasanipalai K. Ananthapadmanabhan, Eddie-Lun Tik Choy, Pengjun Huang, Sharath Manjunath
• Original Assignee: Andrew P. Dejaco, Arasanipalai K. Ananthapadmanabhan, Eddie-Lun Tik Choy, Pengjun Huang, Sharath Manjunath
• Inventors: Choy, Eddie-Lun Tik, Manjunath, Sharath, DeJaco, Andrew P., Ananthapadmanabhan, Arasanipalai K., Huang, Pengjun
• Granted Patent: US 7,426,466 B2
• Time in Patent Office: Days
• US Class Current: 704/219
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/032 Quantisation or dequantisation of spectral components
• CPC Class Codes: G10L 19/04 using predictive techniques
• CPC Class Codes: G10L 19/08 Determination or coding of the excitation function; Determination or coding of the long-term prediction parameters
• CPC Class Codes: G10L 19/097 using prototype waveform decomposition or prototype waveform interpolative [PWI] coders
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 25/12 the extracted parameters being prediction coefficients

Abstract

A method and apparatus for predictively quantizing voiced speech includes a parameter generator and a quantizer. The parameter generator is configured to extract parameters from frames of predictive speech such as voiced speech, and to transform the extracted information to a frequency-domain representation. The quantizer is configured to subtract a weighted sum of the parameters for previous frames from the parameter for the current frame. The quantizer is configured to quantize the difference value. A prototype extractor may be added to first extract a pitch period prototype to be processed by the parameter generator.

Claims

1. A speech coder output frame, comprising:

a quantized target error vector of pitch lag components;

a quantized target error vector of amplitude components;

a quantized target error vector of phase components; and

a quantized target error vector of linear spectral information components, wherein the pitch lag components, amplitude components, phase components, and the linear spectral information components have been extracted from a voiced speech frame.

2. The speech coder output frame of claim 1, wherein the quantized target error vector of pitch lag components is based on a target error vector of pitch lag components ({circumflex over (δ

)}L_m) that is described by a formula;

{circumflex over (L)}_m=L_m−

η

_m1L_m1−

η

_m2L_m2−

. . . −

η

_mNL_mN,

wherein the values L_m1, L_m2 . . . , L_mN are the pitch lags for frames m₁, m₂, . . . , m_N, respectively and the values η

_m1, η

_m2, . . . , η

_mN are weights corresponding to frames m₁, m₂, . . . , m_N, respectively.

3. The speech coder output frame of claim 1, wherein the quantized target error vector of amplitude components is based on a target error vector of amplitude components (δ

A_m) that is described by a formula;

δ

A_m=A_m−

α

_m1^TA_m1−

α

_m2^TA_m2−

. . . −

α

_mN^TA_mN,

wherein the values A_m1, A_m2 . . . , A_mN are a subset of the amplitude vector for frames m₁, m₂, . . . , m_N, respectively, and the values α

_m1^T, α

_m2^T, . . . , α

_mN^T are the transposes of corresponding weight vectors.

4. The speech coder output frame of claim 1, wherein the quantized target error vector of phase components is based on a target error vector of phase components (φ

_m) that is described by a formula;

φ

_m=φ

′

<sub>m−

1</sub>,

wherein φ

′

<sub>m−

1</sub> represent the phases of an extracted prototype.

5. The speech coder output frame of claim 1, wherein the quantized target error vector of linear spectral information components is based on a target error vector of linear spectral information components (T_Mⁿ) that is described by a formula:

6. A method for forming a speech coder output frame, comprising:

quantizing a target error vector of pitch lag components;

quantizing a target error vector of amplitude components;

quantizing a target error vector of phase components; and

quantizing a target error vector of linear spectral information components, wherein the pitch lag components, amplitude components, phase components, and the linear spectral information components have been extracted from a voiced speech frame.