Speech spectrum conversion based on speaker interpolation and multi-functional representation with weighting by radial basis function networks

Speech Communication

Volumn 16, Issue 2, 1995, Pages 139-151

  Iwahashi, Naoto ^a   Sagisaka, Yoshinori ^b  

^a SONY CORPORATION   (Japan)

^b ADVANCED TELECOMMUNICATIONS RESEARCH INSTITUTE INTERNATIONAL   (Japan)

Author keywords

Multiple functional representation; Radial basis function; Speaker adaptation; Speaker interpolation; Speech spectrum conversion; Voice conversion

Indexed keywords

FUNCTIONS; INTERPOLATION; SPECTRUM ANALYSIS; SPEECH; SPEECH PROCESSING;

GRADIENT DESCENT METHOD; RADIAL BASIS FUNCTION NETWORKS; SPEAKER INTERPOLATION; SPEECH SPECTRUM CONVERSION; TRAINING DATA;

SPEECH SYNTHESIS;

EID: 0029251946     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/0167-6393(94)00051-B     Document Type: Article

References (17)

1. A segment-based approach to voice conversion
2. Voice conversion through vector quantization
3. Efficient coding of LPC parameters by temporal decomposition
4. Radial basis functions, multi-variable function interpolation and adaptive networks
5. Voice conversion
6. Methods of converting voice characteristics between male and female by modifying pole parameters
7. Log magnitude approximatinon (LMA) filter
8. Adaptive mixtures of local experts
9. Software for a cascade/parallel formant synthesizer
10. Analysis, synthesis, and perception of voice quality variations among female and male talkers
11. Rapid speaker adaptation using speaker-mixture allophone models applied to speaker-independent speech recognition
12. Quality control of speech by modifying formant frequencies and bandwidth
13. Male and female voice source characteristics: Inverse filtering results
14. Speech-quality control by manipulating formant frequency dynamics
15. A network with multipartitioning units
16. Voice transformation using PSOLA technique
17. Quantization properties of transmission parameters in linear predictive systems