Commit History

  1. refactor based on DDD

    Michael Hu
    5009cb8
  2. fix circular dependency

    Michael Hu
    aaa0814
  3. remove fallback to Dummy TTS

    Michael Hu
    e22e786
  4. replace dummy tts class

    Michael Hu
    ae641cf
  5. fix path

    Michael Hu
    b2b15db

  1. refactor tts module

    Michael Hu
    7495571
  2. add faster whisper installation

    Michael Hu
    37aaac6

  1. refactor(stt): replace whisper with faster-whisper for improved performance⁠

    Michael Hu
    b591083

  1. use unsloth whisper model to reduce the model size

    Michael Hu
    6f92dbc

  1. set correct fallback sequence

    Michael Hu
    6034fea
  2. create fallback flow for tts engines

    Michael Hu
    58d9769
  3. fix

    Michael Hu
    e734196
  4. fix import issue

    Michael Hu
    cb90410
  5. handle dia model not available

    Michael Hu
    a316f58
  6. refactor tts

    Michael Hu
    60bd17d
  7. default to parakeet model

    Michael Hu
    27972f7

  1. refator tts part

    Michael Hu
    3ed3b5a
  2. default to parakeet model as it is much faster

    Michael Hu
    31708ca
  3. support nvdia Parakeet TDT ASR

    Michael Hu
    0ee4f42

  1. fix

    Michael Hu
    e68cae8
  2. fix gradio too many requests exception

    Michael Hu
    451d2b1
  3. handle voice generation failure exception

    Michael Hu
    feebf14
  4. update to set a voice

    Michael Hu
    e3843d2
  5. use kokoro fastAPI server to generate voice

    Michael Hu
    22bd0b9
  6. try to use kokoro api server as backup

    Michael Hu
    5a72681

  1. upgrade streamlit version

    Michael Hu
    724aa35
  2. re-activate dia

    Michael Hu
    d5c9fdf
  3. temporarily disable dia

    Michael Hu
    bb33c13
  4. downgrade streamlit to ensure compatiblity with descript-audio-codec

    Michael Hu
    086a552
  5. try fixing build issue

    Michael Hu
    b697102
  6. revert to earlier version of kokoro because of a incompatiblity issue

    Michael Hu
    ad6b877
  7. Revert "Update README.md"⁠

    Michael Hu
    c549dab
  8. Update README.md

    Verified signature
    5b27125
  9. Switch to use Gradio

    Verified signature
    e2894f6
  10. fix box is not a valid method in gradio issue

    Michael Hu
    6514e96
  11. move to Gradio so we can leverage ZeroGPU

    Michael Hu
    f0248ed
  12. fix runtime dia audio is not generated issue

    Michael Hu
    4a9bb1a
  13. fix runtime dia model issue

    Michael Hu
    2d176f4
  14. try resolve buld issue

    Michael Hu
    501f0b5
  15. try resolve build issue

    Michael Hu
    44fb7ee
  16. enhance logging

    Michael Hu
    91223c9
  17. add missing dac (descript-audio-codec) import

    Michael Hu
    419e343
  18. use dia tts as fallback model if kokoro is not available

    Michael Hu
    7b25fdd
  19. add dia tts model

    Michael Hu
    030c851
  20. add sample dia app file

    Michael Hu
    4b0381b
  21. add gitignore file

    Michael Hu
    875a169
  22. fix init file

    Michael Hu
    28a10f3

  1. add dia tts model. Since dia is not yet released to pypi, we pull in the source directly

    Michael Hu
    9c4b958

  1. chore: add espeak-ng to packages.txt⁠

    Michael Hu
    1a3633a
  2. build: add phonemizer-fork dependency to requirements.txt⁠

    Michael Hu
    672e7e1