chrissewell/whisperX
1
0
Fork 0
mirror of https://github.com/m-bain/whisperX.git synced 2025-07-01 18:17:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #246 from m-bain/v3

Browse Source 
V3
This commit is contained in:
Max Bain
2023-05-13 12:18:09 +01:00
committed by GitHub
parent 46b416296f 9ffb7e7a23
commit d8a2b4ffc9
7 changed files with 18 additions and 119 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
whisperx/alignment.py
View File

@ -261,6 +261,10 @@ def align(
word_text = "".join(word_chars["char"].tolist()).strip()
if len(word_text) == 0:
continue
# dont use space character for alignment
word_chars = word_chars[word_chars["char"] != " "]
word_start = word_chars["start"].min()
word_end = word_chars["end"].max()
word_score = round(word_chars["score"].mean(), 3)

4
whisperx/asr.py
View File

@ -14,7 +14,7 @@ from .vad import load_vad_model, merge_chunks
from .types import TranscriptionResult, SingleSegment
def load_model(whisper_arch, device, compute_type="float16", asr_options=None, language=None,
vad_options=None, model=None):
vad_options=None, model=None, task="transcribe"):
'''Load a Whisper model for inference.
Args:
whisper_arch: str - The name of the Whisper model to load.
@ -31,7 +31,7 @@ def load_model(whisper_arch, device, compute_type="float16", asr_options=None, l
model = WhisperModel(whisper_arch, device=device, compute_type=compute_type)
if language is not None:
tokenizer = faster_whisper.tokenizer.Tokenizer(model.hf_tokenizer, model.model.is_multilingual, task="transcribe", language=language)
tokenizer = faster_whisper.tokenizer.Tokenizer(model.hf_tokenizer, model.model.is_multilingual, task=task, language=language)
else:
print("No language specified, language will be first be detected for each audio file (increases inference time).")
tokenizer = None

7
whisperx/transcribe.py
View File

@ -86,6 +86,11 @@ def cli():
align_model: str = args.pop("align_model")
interpolate_method: str = args.pop("interpolate_method")
no_align: bool = args.pop("no_align")
task : str = args.pop("task")
if task == "translate":
# translation cannot be aligned
no_align = True
return_char_alignments: bool = args.pop("return_char_alignments")
hf_token: str = args.pop("hf_token")
@ -139,7 +144,7 @@ def cli():
results = []
tmp_results = []
# model = load_model(model_name, device=device, download_root=model_dir)
model = load_model(model_name, device=device, compute_type=compute_type, language=args['language'], asr_options=asr_options, vad_options={"vad_onset": vad_onset, "vad_offset": vad_offset},)
model = load_model(model_name, device=device, compute_type=compute_type, language=args['language'], asr_options=asr_options, vad_options={"vad_onset": vad_onset, "vad_offset": vad_offset}, task=task)
for audio_path in args.pop("audio"):
audio = load_audio(audio_path)