mattrguerra/transcription-pipelinemainPrivate on GitHub, published here in full

two-speakers.json

Language JSON870 B33 lines

tests/fixtures/two-speakers.json

  1. {
  2. "source": "LibriSpeech test-clean (Panayotov, Chen, Povey and Khudanpur, 2015), CC BY 4.0, via openslr/librispeech_asr on Hugging Face. Four utterances from two readers, joined with 0.8 s of silence between them.",
  3. "turns": [
  4. {
  5. "speaker": "A",
  6. "start": 0.0,
  7. "end": 5.415,
  8. "text": "FOR A FULL HOUR HE HAD PACED UP AND DOWN WAITING BUT HE COULD WAIT NO LONGER",
  9. "id": "1089-134691-0001"
  10. },
  11. {
  12. "speaker": "B",
  13. "start": 6.215,
  14. "end": 12.14,
  15. "text": "HARANGUE THE TIRESOME PRODUCT OF A TIRELESS TONGUE",
  16. "id": "121-121726-0001"
  17. },
  18. {
  19. "speaker": "A",
  20. "start": 12.94,
  21. "end": 18.115,
  22. "text": "PRIDE AFTER SATISFACTION UPLIFTED HIM LIKE LONG SLOW WAVES",
  23. "id": "1089-134691-0004"
  24. },
  25. {
  26. "speaker": "B",
  27. "start": 18.915,
  28. "end": 23.325,
  29. "text": "ANGOR PAIN PAINFUL TO HEAR",
  30. "id": "121-121726-0002"
  31. }
  32. ]
  33. }