Dupaja
/

Text-to-Speech
Transformers
PyTorch
speecht5
text-to-audio
audio

Commit History

  1. Add handling for fractions and hopefully other number uses

    5f8cefd

  1. Increase max length to 400

    e641a54
  2. Added function to convert digits to written numbers, for handling with SpeechT5

    279f06b
  3. Add inflect to handle numbers

    ea60bf3
  4. Update handler.py

    1e92b35
  5. Rewrite to handle larger text

    c5c6476
  6. Update handler.py

    fc3317e
  7. Update handler.py

    c0208f8
  8. Update requirements.txt

    f503fca
  9. Changes for time tracking

    54df522
  10. Update handler.py

    3b62ecd
  11. Update handler.py

    4fd4826
  12. Added device map

    4577e7e
  13. Update handler.py

    617d2a4
  14. Update handler.py

    9cf2190
  15. Attempt to speed up

    bd9bb8e
  16. Update requirements.txt

    d0c3581
  17. Revert direct file send.

    d16558b
  18. Update handler.py

    910abc8
  19. Try sending wav file directly

    d734a73
  20. Update handler.py

    be0ef80
  21. Remove unused file write

    2349229
  22. Add trust_remote_code=True to dataset

    9594309
  23. Attempt to reduce latency by moving more to init

    9277202

  1. Update handler.py

    0824f3c
  2. Update to personal repos

    5e7a614
  3. Update requirements.txt

    8837630
  4. Updating using SpeechT5 Article⁠

    9f940c9
  5. Update handler.py

    8869945
  6. Update handler.py

    a4c5e15
  7. Update handler.py

    402e7d8
  8. Update handler.py

    9f3bcf5
  9. Update handler.py

    7b5f670
  10. Update requirements.txt

    3c41541
  11. Update requirements.txt

    0b5a163
  12. Update handler.py

    d65f95a
  13. Update requirements.txt

    1ddbd3e
  14. Fixes

    74cf751
  15. Create handler.py

    aaf0168
  16. Update requirements.txt

    df81a21
  17. Update requirements.txt

    bf5e7a6
  18. Create requirements.txt

    215d269
  19. Duplicate from microsoft/speecht5_tts⁠

    Dupaja +1: Matthijs
    4953638