| .. |
|
assets
|
839639a223
Use tiktoken (#1044)
|
2 years ago |
|
normalizers
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 years ago |
|
__init__.py
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 years ago |
|
__main__.py
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 years ago |
|
audio.py
|
919a713499
attempt to fix the repetition/hallucination issue identified in #1046 (#1052)
|
2 years ago |
|
decoding.py
|
c4b50c0824
kwargs in decode() for convenience (#1061)
|
2 years ago |
|
model.py
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 years ago |
|
timing.py
|
255887f219
Squash long words at window and sentence boundaries. (#1114)
|
2 years ago |
|
tokenizer.py
|
b5851c6c40
Update tokenizer.py (#1163)
|
2 years ago |
|
transcribe.py
|
38f2f4d99d
fix all_tokens handling that caused more repetitions and discrepancy in JSON (#1060)
|
2 years ago |
|
triton_ops.py
|
b80bcf610d
apply formatting with `black` (#1038)
|
2 years ago |
|
utils.py
|
c6e4e5efb3
remove auxiliary audio extension (#1021)
|
2 years ago |
|
version.py
|
6dea21fd7f
Release 20230314
|
2 years ago |
ryanheise
255887f219
Squash long words at window and sentence boundaries. (#1114)