Show HN: Real-time local TTS (31M params, 5.6x CPU, voice cloning, ONNX)(github.com)
4 pointsby ZDisket3 days ago3 comments
  • chenglin973 days ago
    Does it support other languages than English? I’m Chinese speaker
    • ZDisket2 days ago
      No multilingual capabilities yet, although that is planned for next iteration.
  • popalchemist3 days ago
    given the architecture, is there a way to force the use of specific phonemes for hard-to-pronounce words? If so that's big
    • ZDisket2 days ago
      Yes. Specifically, the pipeline is text -> phonemizer -> phonemized text -> TTS model -> audio You just have to modify the phonemizer's dictionary.
  • adriencr812 days ago
    [dead]