• perishthethought
    link
    fedilink
    English
    arrow-up
    10
    ·
    5 months ago

    Am I crazy or is 10,000 samples nowhere near enough for training people’s voices?

    • eveninghere@beehaw.org
      link
      fedilink
      arrow-up
      3
      ·
      5 months ago

      If you have pre-trained model or a classical voice matching algorithm as the basis, few samples might suffice.

    • sunzu@kbin.run
      link
      fedilink
      arrow-up
      1
      ·
      5 months ago

      Doubtful that’s enough to do anything useful, maybe if data is great and perfectly tuned with some guidance?

    • Kissaki@beehaw.org
      link
      fedilink
      English
      arrow-up
      1
      ·
      5 months ago

      I don’t think it seems like too few samples for it to work.

      What they train for is rather specific. To identify anger and hostility characteristics, and adjust pitch and inflection.

      Dunno if you meant it like that when you said “training people’s voices”, but they’re not replicating voices or interpreting meaning.

      learned to recognize and modify the vocal characteristics associated with anger and hostility. When a customer speaks to a call center operator, the model processes the incoming audio and adjusts the pitch and inflection of the customer’s voice to make it sound calmer and less threatening.