Speaker Diarization with LSTM: Android Demo

Поділитися
Вставка
  • Опубліковано 16 бер 2019
  • Home page: google.github.io/speaker-id/p...
    Paper: arxiv.org/abs/1710.10468
    Poster: 162.242.252.85/documents/speaker-diarization-lstm
    Tutorial: • [ICASSP 2018] Google's...
    The audios were being played from a speaker, so there were some acoustic distortions.
    I was holding another phone to record the videos with single hand, so the videos are not very stable.
    Udemy online course on speaker recognition: www.udemy.com/course/speaker-...
    Udemy online course on speaker diarization: www.udemy.com/course/diarizat...
  • Наука та технологія

КОМЕНТАРІ • 14

  • @QuanWang
    @QuanWang  2 роки тому

    After years of preparation, I'm excited to share that my online course on Speaker Recognition now starts to accept enrollment on Udemy: www.udemy.com/course/speaker-recognition/?referralCode=1914766AF241CE15D19A
    Please contact me if you need a coupon. Looking forward to seeing you in the lectures!

  • @rohitswami8938
    @rohitswami8938 5 років тому +1

    Absolutely amazing work!

  • @d0vbysh
    @d0vbysh 4 роки тому

    Wow!
    The best example I've ever seen so far!
    Can you share this example via code?

  • @saamermansoor4399
    @saamermansoor4399 Рік тому

    Have you seen anything like this done on iOS using the same principle?

  • @julianasanchez7752
    @julianasanchez7752 3 роки тому

    Is this APP demo available for the public? Would love to give it a try

  • @maulikmadhavi
    @maulikmadhavi 4 роки тому

    Is this offline or online system?

  • @user-yy3nm5hu4g
    @user-yy3nm5hu4g 2 роки тому

    do you need to enroll the speakers' voice first? or it can distinguish the speakers without enrollment process?

    • @QuanWang
      @QuanWang  2 роки тому +1

      Enrollment is not needed for diarization.

    • @user-yy3nm5hu4g
      @user-yy3nm5hu4g 2 роки тому

      ​@@QuanWang Thanks for the response. So with diarization, you can only know "when" the speaker change, but can not know who is speaking? or we can also enhance the function to know "who" and "when"?

    • @QuanWang
      @QuanWang  2 роки тому +1

      @@user-yy3nm5hu4g You know when and who. But this "who" is anonymized. It's like speaker A and speaker B, not Patrick and Mary.

    • @QuanWang
      @QuanWang  2 роки тому +1

      @@user-yy3nm5hu4g Please check the tutorial video.

    • @user-yy3nm5hu4g
      @user-yy3nm5hu4g 2 роки тому

      @@QuanWang Will do! Thanks a lot =)

  • @charanraj1722
    @charanraj1722 4 роки тому

    any one please send the project