‹ BackHN Continuity

Thread

Grok Voice Transcribe 2.0

16 points · 5 comments · vertigoruntime

Loading the complete thread in the background. This saved snapshot is available now. Refresh

  1. bigyabai · · focus · HN ↗
    No weights, no interest.
  2. nomel · · focus · HN ↗
    > Word-level timestamps. Each word has precise start/end times and confidence scores.

    Neat. This is missing from many models.

    1. trentor · · focus · HN ↗
      Because it's a harness feature as it is here.
    2. throwaway89201 · · focus · HN ↗
      Mistral STT offers it, which is an open model.
  3. exizt88 · · focus · HN ↗
    I don't understand how these benchmarks get things so wrong. Nova-3 is by far the best model on the market in WER. Azure's STT is atrocious. And yet every benchmark rates Azure's STT higher than Nova-3's.
    1. mistic92 · · focus · HN ↗
      Mai 2 transcribe is giving me better results than nova
  4. mistic92 · · focus · HN ↗
    If only they had ZDR
  5. xnx · · focus · HN ↗
    $0.10/hour is a good price vs $0.30/hour for Gemini Transcribe 3.5.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.