Senior research scientist in NTT Communication Science Laboratories.
   [More profile]

• Speech enhancement
  1. Statistical signal processing for microphone array control
  2. Speech dereverberation based on maximum likelihood estimation with time-varying Gaussian source model [Sound Demonstrations]
  3. Single channel speech dereverberation based on harmonic structure [Sound Demonstrations]
  4. Sound source separation based on harmonic structure and location of sound source [Sound Demonstrations]
• Robust speech recognition in a real world
  1. Mixed speech recognition [PDF]
• Unified approach to speech recognition, speech synthesis and sound source separation
  1. HMM-STRAIGHT based speech recognition/synthesis system [PDF]
• Robust feature extraction of speech
  1. Fundamental frequency and voiced/unvoiced segments estimation based on dominant harmonic components [PDF]
  2. Fundamental frequency analysis for infant voice [PDF]
• Computational auditory scene analysis. [PhD Thesis]

   E-mail: nak@cslab.kecl.ntt.co.jp
   Voice: +81 774 93 5134
   Fax: +81 774 93 5105
   Tomohiro Nakatani
   Speech Open Lab.
   NTT Communication Science Labs.
   NTT Corporation.
   2-4, Hikaridai, Seika-cho, Soraku-gun, Kyoto 619-0237 Japan