Robust speaking rate estimation using broad phonetic class recognition

Abstract

Robust speaking rate estimation can be useful in automatic speech recognition and speaker identification, and accurate, automatic measures of speaking rate are also relevant for research in linguistics, psychology, and social sciences. In this study we built a broad phonetic class recognizer for speaking rate estimation. We tested the recognizer on a variety of data sets, including laboratory speech, telephone conversations, foreign accented speech, and speech in different languages, and we found that the recognizer's estimates are robust under these sources of variation. We also found that the acoustic models of the broad phonetic classes are more robust than those of the monophones for syllable detection. KeywordsSpeaking rate estimation, syllable detection, robustness, broad phonetic class ©2010 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.This conference paper is available at ScholarlyCommons: http://repository.upenn.edu/cis_papers/444 ROBUST SPEAKING RATE ESTIMATION USING BROAD PHONETIC CLASS RECOGNITION Jiahong Yuan and Mark LibermanUniversity of Pennsylvania ABSTRACTRobust speaking rate estimation can be useful in automatic speech recognition and speaker identification, and accurate, automatic measures of speaking rate are also relevant for research in linguistics, psychology, and social sciences. In this study we built a broad phonetic class recognizer for speaking rate estimation. We tested the recognizer on a variety of data sets, including laboratory speech, telephone conversations, foreign accented speech, and speech in different languages, and we found that the recognizer's estimates are robust under these sources of variation. We also found that the acoustic models of the broad phonetic classes are more robust than those of the monophones for syllable detection.

An Automatic Prolongation Detection Approach in Continuous Speech With Robustness Against Speaking Rate Variations

“…As all syllables contain a vowel, the general idea in speaking rate estimation is based on counting the vowels in speech signals. Signal energy and zero-crossing rate,[18] signal energy and fundamental frequency,[19] and vowel classifier[20] are examples of speaking rate estimation techniques. In this study, we employed signal energy and zero-crossing rate as suggested in Pfau and Ruske[18] to estimate the number of syllables.…”

Evaluation of syllable rate estimation in expressive speech and its contribution to emotion recognition

“…Next, a simple slope based peak counting algorithm was used to get the positions of the syllable nuclei. Yuan and Liberman [24] used a broad phonetic class recognizer for syllable detection and speech rate estimation.…”

A mode-shape classification technique for robust speech rate estimation and syllable nuclei detection

“…The speech rate estimation typically involves identification of the syllable nuclei locations followed by syllable rate computation ( Reddy et al, 2013 ). Generally the approaches for the speech rate estimation and the syllable nuclei detection are based on either acoustic features ( Heinrich and Schiel, 2011;Morgan et al, 1997;Reddy et al, 2013;Wang and Narayanan, 2007 ) or hidden Markov model (HMM) based recognition systems ( Cincarek et al, 2009;Cucchiarini et al, 2000;Hönig et al, 2012;Yuan and Liberman, 2010 ).…”

“…The acoustic feature based methods are typically developed using acoustic properties of the vowels, which in general correspond to the syllable nuclei. Therefore, the vowel rate corresponds directly to the syllable rate ( Pfau and Ruske, 1998;Yuan and Liberman, 2010 ).…”