Demonstration of sound stream segregation

Related Publications

• [List of Papers]

Demonstration sounds appearing at Kagakubu introduction video

• Reference:
  Binaural speech stream segregation with echo (Not published yet)
  
• Segregation from a mixture of two women's voices uttering different Japanese greetings.
  1. Mixture of two women's voices
     
  2. Segregated woman-1's voice
     
  3. Segregated woman-2's voice
     

• Reference:
  Hiroshi G. Okuno, Tomohiro Nakatani, and Takeshi Kawabata: Interfacing Sound Stream Segregation to Speech Recognition Systems --- Preliminary Results of Listening to Several Things at the Same Time. Proc. of AAAI-96, Vol.2., pp.1082--1089, Aug. (1996). [Abstract] [Paper in postscript]
  
• Sound information
  
• Segregation from a mixture of two women's voices uttering different Japanese words.
  1. Original woman-1's voice, "omomuki"
     
  2. Original woman-2's voice, "akichi"
     
  3. Input sound mixture
     
  4. Segregated woman-1's vowels
     
  5. Segregated woman-2's vowels
     
  6. Segregated woman-1's speech
     
  7. Segregated woman-2's speech