Show HN: Nari Qwen3-TTS and Qwen3-ASR – High accuracy, low latency and cost

 (narilabs.com)

12 points | by toebee 2 hours ago

6 comments

  • rahimnathwani 8 minutes ago
    For some reason it switched voices half way through a 33 second clip.

    For OP the clip name is nari-nina-01a0a12f-980a-765e-8029-fa56bd23210d.wav

  • ipsum2 13 minutes ago
    If you're going to announce a TTS model, service, or whatever, you really need demos.
  • asaiacai 48 minutes ago
    This is really cool work! I'm curious like what do you see as the biggest lever for speeding up TTS models or from a technical perspective that this was a promising direction in the first place to push on. If I were to guess, some distillation but I'm certain there are probably TTS model aware architectural changes that just make inference wayyyy faster?
  • meatmanek 30 minutes ago
    > and Qwen3-ASR

    Is the ASR inference engine open source as well?

  • nthypes 12 minutes ago
    [dead]