2 research outputs found
Extrapolation of wideband speech from the telephone band
No full textgrantor:
University of TorontoTelephone speech is bandlimited to the frequency range between 300 and 3300 Hz, which compromises its quality. Wideband speech, accommodating frequencies up to 7000 Hz, provides higher quality but at a cost of increased transmission bandwidth. The proposed pseudo-wideband (PWB) speech algorithm regenerates approximations of the bands missing from telephone speech. This is possible because of the strong inter-band correlations which stem from the acoustics of the production apparatus. For this receiver-based algorithm, the improvement in effective bandwidth requires no extra transmission bandwidth, and involves no codec standardization issues. The spectral envelope and spectral detail are deconvolved via linear predictive analysis, and each is mapped independently to its PWB counterpart. The algorithm is based on parametric analysis using a uniform tube tract model, and has good potential for speaker independence. Performance was encouraging for a preliminary investigation, but a more sophisticated acoustic model is desirable for additional quality improvement.M.A.Sc
Extrapolation of wideband speech from the telephone band
No full textgrantor:
University of TorontoTelephone speech is bandlimited to the frequency range between 300 and 3300 Hz, which compromises its quality. Wideband speech, accommodating frequencies up to 7000 Hz, provides higher quality but at a cost of increased transmission bandwidth. The proposed pseudo-wideband (PWB) speech algorithm regenerates approximations of the bands missing from telephone speech. This is possible because of the strong inter-band correlations which stem from the acoustics of the production apparatus. For this receiver-based algorithm, the improvement in effective bandwidth requires no extra transmission bandwidth, and involves no codec standardization issues. The spectral envelope and spectral detail are deconvolved via linear predictive analysis, and each is mapped independently to its PWB counterpart. The algorithm is based on parametric analysis using a uniform tube tract model, and has good potential for speaker independence. Performance was encouraging for a preliminary investigation, but a more sophisticated acoustic model is desirable for additional quality improvement.M.A.Sc
