Speaking Rate Normalization with Lattice-based Context-dependent Phoneme Duration Modeling for Personalized Speech Recognizers on Mobile Devices

Voice access of cloud applications including social networks using mobile devices becomes attractive today. And personal-ized speech recognizers over mobile devices become feasible because most mobile devices have only a single user. Speak-ing rate variation is known to be an important source of per-formance degradation for spontaneous speech recognition. Speaking rate is speaker dependent, it changes from time to time for every speaker. Furthermore, the speaking rate varia-tion pattern is unique for each speaker. An approach of contin-uous frame rate normalization (CFRN) [1] was recently pro-posed to take care of the speaking rate variation problem. In this paper, we further proposed an extended version of CFRN for personalized speech recognizers on mobile platforms. In this approach, we use context-dependent phoneme duration models adapted to each speaker to estimate the speaking rate utterance by utterance based on lattices obtained with a first-pass recognizer. The proposed approach was evaluated on both read speech and spontaneous recordings from mobile plat-forms and significant improvement were observed in the ex-perimental result. Copyright © 2013 ISCA.