Speech Prosody 2004

Nara, Japan
March 23-26, 2004

Automatic Estimation of Speaking Rate in Multilingual Spontaneous Speech

François Pellegrino (1), J. Farinas (2), J. L. Rouas (2)

(1) Laboratoire Dynamique Du Langage, UMR 5596 CNRS - Univ. Lumière Lyon 2, France
(2) Institut de Recherche en Informatique de Toulouse, UMR 5505 CNRS - Univ. Toulouse 3, France

An automatic estimation of speaking rate is developed in this paper. It is based on an unsupervised vowel detection algorithm and thus may be costlessly applied to any language. Validation is driven on a spontaneous speech subset of the OGI Multilingual Telephone Speech Corpus. The correlation coefficient between the estimated and real speaking rates (evaluated in term of vowel-per-second rates) is 0.84 on average among the 6 languages for which a phonetic transcription is available (English, German, Hindi, Japanese, Mandarin and Spanish).

Full Paper

Bibliographic reference.  Pellegrino, François / Farinas, J. / Rouas, J. L. (2004): "Automatic estimation of speaking rate in multilingual spontaneous speech", In SP-2004, 517-520.