A Multi-Space Distribution (MSD) and two-stream tone modeling approach to Mandarin speech recognition

@article{Qian2009AMD,
  title={A Multi-Space Distribution (MSD) and two-stream tone modeling approach to Mandarin speech recognition},
  author={Yao Qian and Frank K. Soong},
  journal={Speech Communication},
  year={2009},
  volume={51},
  pages={1169-1179}
}
Tone plays an important role in recognizing spoken tonal languages like Chinese. However, the discontinuity of F0 between voiced and unvoiced transition has traditionally been a hurdle in creating a succinct statistical tone model for automatic speech recognition and synthesis. Various heuristic approaches have been proposed before to get around the problem but with limited success. The Multi-Space Distribution (MSD) proposed by Tokuda et al. which models the two probability spaces, discrete… CONTINUE READING

Results and Topics from this paper.

Key Quantitative Results

  • Comparing with the conventional system where F0 contours are interpolated in unvoiced segments, our approach improves the recognition performance by 9.8%, 7.4% and 13.3% in relative TSER reductions in the corresponding speech recognition tasks, respectively.

Citations

Publications citing this paper.
SHOWING 1-5 OF 5 CITATIONS

A DNN-based acoustic modeling of tonal language and its application to Mandarin pronunciation training

  • 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2014
VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS

Automatic phonetic segmentation in Mandarin Chinese: Boundary models, glottal features and tone

  • 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2014
VIEW 11 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Improved tone modeling by exploiting articulatory features for mandarin speech recognition

  • 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2012
VIEW 1 EXCERPT
CITES BACKGROUND