Blog(1)

• [INTERSPEECH 2022 Series #5] Prototypical Speaker-Interference Loss for Target Voice Separation Using Non-Parallel Audio Samples

  Deep learning based audio signal processing is increasingly becoming popular for solving the cocktail party problem [1-5]. A noisy signal, which is mixed by a clean and noise signal, is used as a pair with the clean signal for training a speech enhancement or source separation model [6-7].

Research Areas(0)

Publications(1)

• Prototypical speaker-interference loss for target voice separation using non-parallel audio samples

  AuthorSeongkyu Mun, Dhananjaya Gowda, Jihwan Lee, Changwoo Han, Dokyun Lee, Chanwoo Kim

  PublishedAnnual Conference of the International Speech Communication Association (INTERSPEECH)

  Date2022-09-18

News(1)

• [Blog] [INTERSPEECH 2022 Series #5] Prototypical Speaker-Interference Loss for Target Voice Separation Using Non-Parallel Audio Samples

  Deep learning based audio signal processing is increasingly becoming popular for solving the cocktail party problem [1-5]. A noisy signal, which is mixed by a clean and noise signal, is used as a pair with the clean signal for training a speech enhancement or source separation model [6-7].

Others(0)