Voice analysis-synthesis method using noise having diffusion which varies with frequency band to modify predicted phases of transmitted pitch data blocks

• US 5,878,388 A
• Filed: 06/09/1997
• Issued: 03/02/1999
• Est. Priority Date: 03/18/1992
• Status: Expired due to Term

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Ono, Shinobu, Nishiguchi, Masayuki, Matsumoto, Jun
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Chawan, Vijay B.
• Application Number:
  US08/871,812
• Time in Patent Office: 631 Days
• Field of Search: 704/222, 704/207, 704/214, 704/208, 704/205, 704/204, 704/219, 704/233, 704/229, 704/220, 704/221, 341/50, 341/51, 378/253, 378/240
• US Class Current: 704/214
• CPC Class Codes: G10L 19/0212 using orthogonal transformation
• CPC Class Codes: G10L 19/038 Vector quantisation, e.g. TwinVQ audio
• CPC Class Codes: G10L 19/04 using predictive techniques
• CPC Class Codes: G10L 19/10 the excitation function being a multipulse excitation
• 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 2019/0005 Multi-stage vector quantisation
• CPC Class Codes: G10L 2025/937 Signal energy in various frequency bands
• CPC Class Codes: G10L 25/27 characterised by the analysis technique
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals
• CPC Class Codes: G10L 25/93 Discriminating between voiced and unvoiced parts of speech signals G10L25/90 takes precedence

Abstract

A high efficiency encoding method for encoding data on frequency axis obtained by dividing an input audio signal on block-by-block basis and converting the signal onto the frequency axis, wherein V bands are searched for a band B_VH with the highest center frequency if it is decided that there are one or more shift points of voiced (V)/unvoiced (UV) decision data of all bands on the frequency axis, and wherein the number of V bands N_V up to the band B_VH is found, so as to decide whether proportion of the V bands is equal to or higher than a predetermined threshold N_th, thereby deciding one V/UV boundary point. Thus, it is possible to replace the V/UV decision data for each band by information on one demarcation in all bands, thereby to reduce data volume and to reduce bit rate. Also, by using two-stage hierarchical vector quantization in quantizing the data on the frequency axis, operation volume for codebook search and memory capacity of the codebook are reduced.

Claims

1. A voice analysis-synthesis method, comprising the steps of:

dividing an input voice signal on a block-by-block basis and extracting pitch data from each block;

converting the voice signal, on the block-by-block basis, into frequency-domain data;

dividing the frequency-domain data for each of the blocks into plural bands of data on the basis of the pitch data, each of said bands corresponding to a different range of frequencies;

finding power information for each of the bands of said each of the blocks and voiced/unvoiced decision information for said each of the bands of said each of the blocks;

transmitting the pitch data, the power information for said each of the bands of said each of the blocks, and the voiced/unvoiced decision information for said each of the bands of said each of the blocks;

receiving the pitch data, the power information, and the voiced/unvoiced decision information, and predicting a block terminal edge phase for each block of the received pitch data on the basis of said each block of the received pitch data and a block initial phase for said each block of the received pitch data; and

modifying the predicted block terminal edge phase, using noise having diffusion which varies from band to band for each of the bands.

2. The voice analysis-synthesis method as claimed in claim 1, wherein the noise is Gaussian noise.

3. A pitch extraction method for processing an input audio signal comprising frames, each of the frames corresponding to a different time along a time axis, said method comprising the steps of:

detecting plural peaks from auto-correlation data of a current frame, where the current frame is one of said frames; and

detecting a pitch of the current frame by determining a position of a maximum peak among the detected plural peaks of the current frame when the maximum peak is equal to or larger than a predetermined threshold, and deciding the pitch of the current frame by determining a position of a peak in a pitch range having a predetermined relation with a pitch found in one of the frames other than said current frame when the maximum peak is smaller than the predetermined threshold.