A new method is proposed for automatically evaluating the English pronunciation quality of Japanese speakers. It is assumed that this can be determined based on three criteria: the static characteristics of phonetic spectra, the dynamic structure of spectrum sequences, and the prosodic characteristics of utterances. Evaluation is carried out by comparing English words pronounced by a Japanese with those pronounced by a native speaker using speech recognition techniques. Preliminary experiments show the evaluation results obtained using the proposed method to correspond well with human judgement of pronunciation quality.
Bibliographic reference. Hamada, Hiroshi / Miki, Satoshi / Nakatsu, Ryohei (1989): "Automatic evaluation of English pronunciation based on speech recognition techniques", In EUROSPEECH-1989, 1421-1424.