Skip to content

Latest commit

 

History

History
86 lines (66 loc) · 5.99 KB

CHANGELOG.md

File metadata and controls

86 lines (66 loc) · 5.99 KB

CHANGELOG

  • Relax triton requirements for compatibility with pytorch 2.1 and newer (#1802)
  • remove tiktoken pin (#1759)
  • docs: Disambiguation of the term "relative speed" in the README (#1751)
  • allow_pickle=False while loading of mel matrix IN audio.py (#1511)
  • handling transcribe exceptions. (#1682)
  • Add new option to generate subtitles by a specific number of words (#1729)
  • Fix exception when an audio file with no speech is provided (#1396)
  • Add .pre-commit-config.yaml (#1528)
  • fix doc of TextDecoder (#1526)
  • Update model-card.md (#1643)
  • word timing tweaks (#1559)
  • Avoid rearranging all caches (#1483)
  • Improve timestamp heuristics. (#1461)
  • fix condition_on_previous_text (#1224)
  • Fix numba depreceation notice (#1233)
  • Updated README.md to provide more insight on BLEU and specific appendices (#1236)
  • Avoid computing higher temperatures on no_speech segments (#1279)
  • Dropped unused execute bit from mel_filters.npz. (#1254)
  • Drop ffmpeg-python dependency and call ffmpeg directly. (#1242)
  • Python 3.11 (#1171)
  • Update decoding.py (#1219)
  • Update decoding.py (#1155)
  • Update README.md to reference tiktoken (#1105)
  • Implement max line width and max line count, and make word highlighting optional (#1184)
  • Squash long words at window and sentence boundaries. (#1114)
  • python-publish.yml: bump actions version to fix node warning (#1211)
  • Update tokenizer.py (#1163)
  • abort find_alignment on empty input (#1090)
  • Fix truncated words list when the replacement character is decoded (#1089)
  • fix github language stats getting dominated by jupyter notebook (#1076)
  • Fix alignment between the segments and the list of words (#1087)
  • Use tiktoken (#1044)
  • kwargs in decode() for convenience (#1061)
  • fix all_tokens handling that caused more repetitions and discrepancy in JSON (#1060)
  • fix typo in CHANGELOG.md
  • Fix the repetition/hallucination issue identified in #1046 (#1052)
  • Use triton==2.0.0 (#1053)
  • Install triton in x86_64 linux only (#1051)
  • update setup.py to specify python >= 3.8 requirement
  • remove auxiliary audio extension (#1021)
  • apply formatting with black, isort, and flake8 (#1038)
  • word-level timestamps in transcribe() (#869)
  • Decoding improvements (#1033)
  • Update README.md (#894)
  • Fix infinite loop caused by incorrect timestamp tokens prediction (#914)
  • drop python 3.7 support (#889)
  • handle printing even if sys.stdout.buffer is not available (#887)
  • Add TSV formatted output in transcript, using integer start/end time in milliseconds (#228)
  • Added --output_format option (#333)
  • Handle XDG_CACHE_HOME properly for download_root (#864)
  • use stdout for printing transcription progress (#867)
  • Fix bug where mm is mistakenly replaced with hmm in e.g. 20mm (#659)
  • print '?' if a letter can't be encoded using the system default encoding (#859)

The first versioned release available on PyPI