This paper presents a linear prediction (LP) based vocoder in which speech waveforms are considered as having a 'time envelope', the shape of which contains important perceptual information. By ensuring that the time envelope of the synthetic speech closely matches that of the original, natural sounding synthetic speech can be produced. The advantage over more traditional linear prediction vocoders is that the amplitude time envelope is preserved in addition to the spectral envelope, allowing the rapid amplitude transitions associated with onsets to be retained in the synthetic speech, resulting in a more intelligible output. This paper presents a complete vocoder scheme including details of techniques such as parameter interpolation, quantisation, spectrum shaping and pitch detection which have proven necessary to produce natural sounding synthetic speech.
Bibliographic reference. Atkinson, I. A. / Kondoz, A. M. / Evans, B. G. (1995): "Time envelope LP vocoder: a new coding technique at very low bit rates", In EUROSPEECH-1995, 241-244.