포만트 공간에서의 주파수 변환을 이용한 이중 언어 음성 변환 연구
Bilingual Voice Conversion Using Frequency Warping on Formant Space
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
This paper describes several approaches to transform a speaker’s individuality to another’s individuality using frequencywarping between bilingual formant frequencies on different language environments. The proposed methods are simple andintuitive voice conversion algorithms that do not use training data between different languages. The approaches find thewarping function from source speaker’s frequency to target speaker’s frequency on formant space. The formant spacecomprises four representative monophthongs for each language. The warping functions can be represented by piecewise linearequations, inverse matrix. The used features are pure frequency components including magnitudes, phases, and line spectralfrequencies (LSF). The experiments show that the LSF-based voice conversion methods give better performance than othermethods.
Abstract
This paper describes several approaches to transform a speaker’s individuality to another’s individuality using frequencywarping between bilingual formant frequencies on different language environments. The proposed methods are simple andintuitive voice conversion algorithms that do not use training data between different languages. The approaches find thewarping function from source speaker’s frequency to target speaker’s frequency on formant space. The formant spacecomprises four representative monophthongs for each language. The warping functions can be represented by piecewise linearequations, inverse matrix. The used features are pure frequency components including magnitudes, phases, and line spectralfrequencies (LSF). The experiments show that the LSF-based voice conversion methods give better performance than othermethods.
참고문헌 (17)
- Mizuno, H., "Voice conversion algorithm based on peicewise linear conversion rules of formant frequency and spectrum tilt", Speech Communication, 16, pp.153~164 (1995)
- Kuwabara H., "Acoustic characteristics of speaker individuality: Control and conversion", Speech Communication, 16, pp.165~173 (1995)
- Narendranath M., "Transformation of foramnts for voice conversion using artificial neural networks", Speech Communication, 16, pp.207~216 (1995)
- Sundermann D., "Time domain vocal tract length normalization", Proc. of IEEE In. Symposium on Signal Processing and Information Technology, pp.191~194 (2004)
- Errno D., "Voice conversion based on weighted frequency warping", IEEE Tr. on Audio, Speech, and Language Processing, 18(5), pp.922~1931 (2010)
- Pye D., "Experiments in speaker normalization and adaptation for large vocabulary speech recognition", Proc. of IEEE Int. Conference on Acoustics, Speech and Signal Processing, pp.1047~1050 (1997)
- Sundermann D., "VTLN-Based cross-language voice conversion", Proc. of IEEE Automatic Speech Recognition and Understanding Workshop, pp.676~681 (2003)
- Saheer L., "Vocal tract length normalization for statistical parametric speech synthesis", IEEE Tr. on Audio, Speech, and Language Processing, 20(7), pp.2134~2148 (2012)
- Sundermann D., "Text-independent voice conversion based on unit selection", Proc. of Int. Conference on Acoustics, Speech and Signal Processing, 1, pp.81~84 (2006)
- Huang X., "Spoken language processing – A guide to theory, algorithm, and system development" (2001), Prentice Hall
- Yun Y. -S., "Bilingual voice conversion by weighted frequency warping based on formant space", LNCS, 8082, pp.137~144 (2013)
- Sundermann D., "Evaluation of VTLN-Based voice conversion for embedded speech synthesis", Int. Proc. of Conference on Spoken Language Processing, pp.3~6 (2005)
- Erro D., "Voice conversion based on weighted frequency warping", IEEE Tr. on Audio, Speech, and Language Processing, 20(7), pp.2134~2148 (2010)
- 윤영선, "주파수 직접 변환에 의한 다국어 음성 변환 연구", 2013 한국음성학회 가을 학술대회 발표 논문집, pp.127~128 (2013)
- "VoiceTextTM" (2014)[URL]
- "Jennifer Clyde Interview" (2014)[URL]
- Fant G., "Acoustic theory of speech production" (1970), The Hague