Abstract

The whisper conversion technology is to transform undistinguished whispers with lower SNR into clear normal speech, and it has important application prospect in mobile communication. Because the whisper speech is stirred by the yawp source, its formant position shifts and its bandwidth increases, which induces the problem of formant combination occurs in the whisper conversion. By analyzing the power spectrum, in this article, we proposed a modified adaptive algorithm for formant bandwidth. Based on the rule that the pole power does not change, the algorithm has resolved the problem of formant combination by modifying the formant bandwidth of whisper before implementing formant conversion. The experimental results with six Chinese mandarin monophthong phoneme conversions proved the validity of the algorithm.

Highlights

  • (McCandless S, 1974, P.135-141), and Figure 2 is the formant of the normal speech [o4] after conversion by the Gauss mixture model (Lv, 2004)

  • Based on the rule that the pole power does not change, in this article, we propose the algorithm which automatically add the energy of the deleted formant to the reserved formant and realize the adaptive formant bandwidth change

  • The point lineation is three formants obtained by traditional LPC algorithm, and the thin real line is the normal speech frequency spectrum curve after conversion obtained by Gauss mixture model, and from the figure, we can intuitively see the combination of formants F1 and F2

Read more

Summary

Introduction

(McCandless S, 1974, P.135-141), and Figure 2 is the formant of the normal speech [o4] after conversion by the Gauss mixture model (Lv, 2004). 2. Formant combination in whisper conversion Figure 1 is the former three formants extracting whisper signal [o4] by the linear predictive spectrum algorithm

Results
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.