 Guillaume Klein
|
671ac5a4ce
Fix alignment between the segments and the list of words (#1087)
|
2 gadi atpakaļ |
 Jong Wook Kim
|
839639a223
Use tiktoken (#1044)
|
2 gadi atpakaļ |
 Jong Wook Kim
|
ad3250a846
Release 20230308
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
c4b50c0824
kwargs in decode() for convenience (#1061)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
38f2f4d99d
fix all_tokens handling that caused more repetitions and discrepancy in JSON (#1060)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
aac47c9834
fix typo
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
26807ec6d3
Release 20230307
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
919a713499
attempt to fix the repetition/hallucination issue identified in #1046 (#1052)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
38e990d853
Use triton==2.0.0 (#1053)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
924e1f8e06
Try installing triton only if linux & x86_64 (#1051)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
4b0d5e58d0
Update setup.py
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
8180fde939
Release 20230306
|
2 gadi atpakaļ |
 Local State
|
c6e4e5efb3
remove auxiliary audio extension (#1021)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
500d0fe966
word-level timestamps in `transcribe()` (#869)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
eab8d920ed
Decoding improvements (#1033)
|
2 gadi atpakaļ |
 Roman Vasilenko
|
3e1780fd37
Update README.md (#894)
|
2 gadi atpakaļ |
 Andrey Chernykh
|
7858aa9c08
Fix infinite loop caused by incorrect timestamp tokens prediction (#914)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
5c1a8c10e7
clarify that 3.11 is not supported
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
4e635c6644
Update README.md about Python 3.8+ requirement
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
a6b36ede1f
drop python 3.7 support (#889)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
55f690af79
Release 20230124
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
7f1ef223ab
handle printing even if sys.stdout.buffer is not available (#887)
|
2 gadi atpakaļ |
 Niels Mayer
|
f5bfe004ec
Add TSV formatted output in transcript, using integer start/end times in milliseconds. (#228)
|
2 gadi atpakaļ |
 Aaryan YVS
|
da600abd2b
Added --output_format option (#333)
|
2 gadi atpakaļ |
 zer0-x
|
9f7aba6099
Handle XDG_CACHE_HOME properly for download_root (#864)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
12e1089462
use stdout for printing transcription progress (#867)
|
2 gadi atpakaļ |
 Markus Hennerbichler
|
ea1c266709
Fix bug where mm is mistakenly replaced with hmm in e.g. 20mm (#659)
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
8135a7c31c
verbose outputs from pytest
|
2 gadi atpakaļ |
|
Jong Wook Kim
|
9d646db9d8
print '?' if a letter can't be encoded using the system default encoding (#859)
|
2 gadi atpakaļ |
