This group works towards automatic transcription of continuous speech with a vocabulary and syntax as unrestricted as possible. It is a long-term effort; however, an experimental system is operational. The acoustic processor contains a spectrum analyzer based on the Fast Fourier Transform and a phone segmenter/recognizer which makes use of transitional and steady-state information in its classification. The linguistic processor accepts an imperfect string of phones and produces an estimated transcription of the speech input.

Continuous speech recognition

Acoustic-phonetic conversion is probably the most critical step in continuous speech recognition. The transitional information can be used as follows, in order to improve the results. First we contitute a lexicon of the phoneme steady-state spectra and a lexicon of all the transitions (diphones), each one being characterized by a"differential spectrum". The unknown continuous speech wave is segmented into quasi steady-state and transitional segments ; the labelling of the quasi steady-state segments admits several candidates. The transitional segment between two quasi steady-state spectra is then compared to the diphones of the lexicon selected from the combination of the surrounding possible phoneme labels. Actually, only the comparisons which are compatible with the recent past of the message are made. When working as a phoneme-vocoder, the whole procedure needs about 3x real-time, without any optimization.

http://ieeexplore.ieee.org/iel6/8365/26347/01170171.pdf

Acoustic-phonetic recognition of connected speech using transient information

The problems involved in automatic recognition of continuous speech are generally considered very difficult. In an attempt to solve these problems an acoustic processing system has been under development which has stages dedicated to signal analysis, spectral classification, phonemic segmentation, phonemic (steady state) classification, phoneme boundary placement, diadic (transitional) classification, and final phoneme string consolidation. This paper presents the concepts of and some details concerning the first five stages, through phoneme boundary placement. Performance results are also presented.

The 1976 modular acoustic processor (MAP) : Signal analysis and phonemic segmentation

We present here several points concerning speech communication with computers, and dealing with different aspects of the problem. We do not intend to write about a definite part of a speech communication system, but rather to give in formation and to report experiments that may be useful for people working in that field.

Some Points Concerning Speech Communication with Computers

A probabilistic vector model was developed to identify phonemically equivalent intervocalic stop consonants independently of the succeeding vowel's identity or the differences among talkers. Acoustic features of stop perception and production were studied as random processes whose probability density functions provided models for describing either the voicing mode or place-of-articulation of a stop. Each acoustic feature present in a stop's production contributed a vector whose magnitude and direction were determined from these acoustically based feature models. Combining these vectors resulted in the stop's identification. The voicing mode model performed as well as trained listeners (greater than 99% correct recognition) while the place-of-articulation model performed 15% below the level of trained listeners.

A probabalistic vector model for identification of intervocalic stop consonants

The Modular Acoustic Processor (MAP), a complex experimental system for automatic derivation of phonemic string output for continuous speech, was first described in April 1974. Many of the new concepts currently in MAP are described. In particular, the problems concerning the diadic segment classification and final string estimation are discussed, and the current solutions given. Results on a large body of continuous speech data, prepared by an automatic evaluation system will also be presented.

N. Dixon

Papers

The 1976 modular acoustic processor (MAP) : Diadic segment classification and final phonemic string estimation