Trajectory Mixture Density Networks with Multiple Mixtures for Acoustic-Articulatory Inversion

@inproceedings{Richmond2007TrajectoryMD,
  title={Trajectory Mixture Density Networks with Multiple Mixtures for Acoustic-Articulatory Inversion},
  author={Korin Richmond},
  booktitle={NOLISP},
  year={2007}
}
We have previously proposed a trajectory model which is based on a mixture density network (MDN) trained with target variables augmented with dynamic features together with an algorithm for estimating maximum likelihood trajectories which respects the constraints between those features. In this paper, we have extended that model to allow diagonal covariance matrices and multiple mixture components in the trajectory MDN output probability density functions. We have evaluated this extended model… CONTINUE READING

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 35 CITATIONS

Preliminary inversion mapping results with a new EMA corpus

  • INTERSPEECH
  • 2009
VIEW 7 EXCERPTS
CITES RESULTS, METHODS & BACKGROUND

Acoustic-Articulatory Modelling with the Trajectory HMM

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Deep acoustic-to-articulatory inversion mapping with latent trajectory modeling

  • 2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)
  • 2017
VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Comparison of HMM and TMDN methods for lip synchronisation

VIEW 10 EXCERPTS
CITES RESULTS & BACKGROUND

Modelling Speech Dynamics with Trajectory-HMMs

VIEW 3 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Acoustic-Articulatory Modeling With the Trajectory HMM

  • IEEE Signal Processing Letters
  • 2008
VIEW 3 EXCERPTS
CITES METHODS
HIGHLY INFLUENCED

Articulatory and Spectrum Information Fusion Based on Deep Recurrent Neural Networks

  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • 2019
VIEW 1 EXCERPT
CITES METHODS

References

Publications referenced by this paper.
SHOWING 1-10 OF 17 REFERENCES

Speech parameter generation algorithms for HMM-based speech synthesis

  • 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100)
  • 2000
VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Speech production knowledge in automatic speech recognition.

  • The Journal of the Acoustical Society of America
  • 2007
VIEW 1 EXCERPT

The MOCHA-TIMIT articulatory database

A. Wrench
  • http://www.cstr.ed.ac.uk/artic/mocha.html
  • 1999
VIEW 1 EXCERPT