Improvement of the narrowband LPC synthesis

Kang, G.; Everett, Stephanie S.

doi:10.1109/icassp.1984.1172445

Cited by 4 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both speakers and listeners were native Americans. For reference purposes, the intelligibility score of currently available LPC vocoders operating at 2,4 kbps, as reported in [10], is 84%. Furthermore, the intelligibility scores of currently available CVSD adaptive delta modulation codecs operating at 32 and 16 kbps are, respectively, 95% and 91%.…”

mentioning

confidence: 99%

Voice-excited predictive coder (VEPC) implementation on a high-performance signal processor

et al. 1985

View full text Add to dashboard Cite

mentioning

confidence: 99%

Voice-excited predictive coder (VEPC) implementation on a high-performance signal processor

et al. 1985

View full text Add to dashboard Cite

Extended glottal LPS with mixed excitation

Bergström

Hedelin

8th European Conference on Electrotechnics, Conference Proceedings on Area Communication

View full text Add to dashboard Cite

A glottal LPC speech coder was proposed by Hedelin [l,2]. which. in the modelling sense. is an extension to the conventional LPC (3.41. For voiced sounds the white source of classical LPC is replaced by a parametric glottal source, consisting of short sinusoidal segments. However. the binary voicedhnvoiced decision remains.In order to make this coder more robust. and to improve the ability of reproducing plosives. voiced fricatives. etc.. we here propose an extension. The binary voicing decision is removed. and the pulse-source is generalized. allowing impulses as well as glottal pulses. Moreover. the timing of the pulses are preserved and used in the synthesis. This pulse source always nins in parallell with a white noise source.The analysis is pelformed hy a step-by-step procedure. Firstly. the production filter and the glottal pulses are estimated using the method described in 121. Secondly. the impulse source parameters are estimated with a standard multi-pulse technique 171. Thirdly. the power of the random source is estimated from the model error.Speech analysed and synthesized using this technique is considered as being of high quality in informal listening tests. As compared to the glottal LPC model. plosives. voiced fricatives. etc. are reproduced better.The computational hurden is heavy. and the input speech must be recorded under good conditions. These limitations imply a possible use in applications such as speech-storing. etc.

show abstract

A new mixed excitation LPC vocoder

McCree

Barnwell

1991

[Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

Improvement of the narrowband LPC synthesis

Cited by 4 publications

References 2 publications

Voice-excited predictive coder (VEPC) implementation on a high-performance signal processor

Voice-excited predictive coder (VEPC) implementation on a high-performance signal processor

Extended glottal LPS with mixed excitation

A new mixed excitation LPC vocoder

Contact Info

Product

Resources

About