Prosody for Mandarin Speech Recognition:a Comparative Study of Read and Spontaneous Speech

摘要

In this paper, we present a comparative study between sponta-neous speech and read Mandarin speech in the context of au-tomatic speech recognition. We focus on analysis and mod-eling of prosodic features, based on a unique speech corpus that contains similar amounts of read and spontaneous speech data from the same group of speakers. Statistical analysis is carried out on tone contours and duration of syllable and sub-syllable units. Speech recognition experiments are performed to evaluate the effectiveness of different approaches to incorpo-rate prosodic features into acoustic modeling. A key problem being addressed is how to deal with the unvoiced frames where FO values are unavailable. We apply the technique of Multi-space distribution (MSD) to model partially continuous FO con-tours. For spontaneous speech, the tonal-syllable error rate is reduced from the MFCC baseline of 64.8% to 59.4% with the MSD based prosody model. For read speech, the performance improves from 46.0% to 36.4%.