Hugging Face's logo

ambassadia
/
supertonic-3-mlx

Text-to-Speech
MLX
Supertonic
supertonic-3-mlx
supertonic-3
apple-silicon
tts
speech-synthesis
multilingual
flow-matching
Model card Files
xet
 Community
supertonic-3-mlx / samples

Commit History

  1. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    b434743

  2. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    ea6316b

  3. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    437708c

  4. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    00270e5

  5. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    f6c119d

  6. quality: VE style_key fix — closes 500-800 Hz centroid gap on dark voices

    ambassadia
    Verified signature
    4bfe13f

  7. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    c81cdbe

  8. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    91700cd

  9. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    7581f08

  10. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    53e6c68

  11. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    9470605

  12. fix: DP residual + re-gen samples (Whisper validation 86% overlap)

    ambassadia
    Verified signature
    e78c26c

  13. v0.1.0 — initial release (MLX-native Supertonic 3, ~x100 RTF on M4)

    ambassadia
    Verified signature
    69dee76