Commit History

Autor SHA1 Mensaxe Data
  Arthur Kim 8b330df096 Add .pre-commit-config.yaml (#1528) hai 1 ano
  sqhao 21010ef454 fix doc of TextDecoder (#1526) hai 1 ano
  Nino Risteski 29b7df6231 Update model-card.md (#1643) hai 1 ano
  taylorchu e8622f9afc word timing tweaks (#1559) hai 1 ano
  WangChou Lu b91c907694 Avoid rearranging all caches (#1483) hai 1 ano
  ryanheise f572f2161b Improve timestamp heuristics. (#1461) hai 1 ano
  Valentin Berkes 248b6cb124 fix condition_on_previous_text (#1224) hai 1 ano
  Paul Willot 7ca9fbea86 Fix numba depreceation notice (#1233) hai 1 ano
  Brett Balquist b1c0815c79 Updated README.md to provide more insight on BLEU and specific appendices (#1236) hai 1 ano
  Théo BOYER e334ff141d Avoid computing higher temperatures on no_speech segments (#1279) hai 1 ano
  petterreinholdtsen 5523722842 Dropped unused execute bit from mel_filters.npz. (#1254) hai 1 ano
  petterreinholdtsen 8035e9ef48 Drop ffmpeg-python dependency and call ffmpeg directly. (#1242) hai 1 ano
  Johnny e69930cb9c Python 3.11 (#1171) hai 1 ano
  Jong Wook Kim c09a7ae299 Update decoding.py (#1219) hai 1 ano
  Fernando O. Gallego b0022b3283 Update decoding.py (#1155) hai 1 ano
  Arseniy Bushyn 76c901ab8d Update README.md to reference tiktoken (#1105) hai 1 ano
  ryanheise 43940fc978 Implement max line width and max line count, and make word highlighting optional (#1184) hai 1 ano
  ryanheise 255887f219 Squash long words at window and sentence boundaries. (#1114) hai 1 ano
  K.B.Dharun Krishna a151816b6b python-publish.yml: bump actions version to fix node warning (#1211) hai 1 ano
  Jong Wook Kim b5851c6c40 Update tokenizer.py (#1163) hai 1 ano
  Jong Wook Kim 6dea21fd7f Release 20230314 %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim 79c43e4859 abort find_alignment on empty input (#1090) %!s(int64=2) %!d(string=hai) anos
  Guillaume Klein 5f9ac653b7 Fix truncated words list when the replacement character is decoded (#1089) %!s(int64=2) %!d(string=hai) anos
  Akash Mahajan ba88b8e1b3 fix github language stats getting dominated by jupyter notebook (#1076) %!s(int64=2) %!d(string=hai) anos
  Guillaume Klein 671ac5a4ce Fix alignment between the segments and the list of words (#1087) %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim 839639a223 Use tiktoken (#1044) %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim ad3250a846 Release 20230308 %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim c4b50c0824 kwargs in decode() for convenience (#1061) %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim 38f2f4d99d fix all_tokens handling that caused more repetitions and discrepancy in JSON (#1060) %!s(int64=2) %!d(string=hai) anos
  Jong Wook Kim aac47c9834 fix typo %!s(int64=2) %!d(string=hai) anos