Research Interest

This work is primarily undertaken at Toshiba Research Europe Ltd.

• Parametric acoustic models for speech synthesis
• Trainable excitation models
• Use of found data for speech synthesis
• Controllable speech synthesis (both speaker and expressive state)

Selected Publications

• Braunschweiler N., Gales M.J.F. and Buchholz S. (2010)
  Lightly supervised recognition for automatic alignment of large coherent speech recordings.
  InterSpeech 2010
• Pilkington N., Zen H. and Gales M.J.F. (2011)
  Gaussian process experts for voice conversion.
  InterSpeech 2011
• Zen, H., Gales M.J.F., Nankaku, Y. and Tokuda, K. (2011)
  Product of Experts for Statistical Parametric Speech Synthesis.
  IEEE Transactions Audio Speech and Language Processing.
• Zen, H., Braunschweiler, N., Buchholz, S.,Gales M.J.F., Knill, K., Krstulovic, S. and Latorre, J. (2012)
  Statistical Parametric Speech Synthesis Based on Speaker and Language Factorization.
  IEEE Transactions Audio Speech and Language Processing.
• Wan V., Latorre, J., Chin K.K., Chen L. and Gales M.J.F., Zen H., Knill K. and Akamine M. (2012)
  Combining multiple high quality corpora for improving HMM-TTS.
  To appear InterSpeech 2012