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

suppress numerals fix

Browse Source
This commit is contained in:
Max Bain
2023-06-05 15:33:04 +01:00
parent b026407fd9
commit 74a00eecd7
1 changed files with 1 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
whisperx/asr.py
View File

@ -11,7 +11,7 @@ from transformers.pipelines.pt_utils import PipelineIterator
from .audio import N_SAMPLES, SAMPLE_RATE, load_audio, log_mel_spectrogram from .audio import N_SAMPLES, SAMPLE_RATE, load_audio, log_mel_spectrogram
from .vad import load_vad_model, merge_chunks from .vad import load_vad_model, merge_chunks
<<<<<<< HEAD from .types import TranscriptionResult, SingleSegment
def find_numeral_symbol_tokens(tokenizer): def find_numeral_symbol_tokens(tokenizer):
numeral_symbol_tokens = [] numeral_symbol_tokens = []
@ -21,12 +21,6 @@ def find_numeral_symbol_tokens(tokenizer):
numeral_symbol_tokens.append(i) numeral_symbol_tokens.append(i)
return numeral_symbol_tokens return numeral_symbol_tokens
def load_model(whisper_arch, device, compute_type="float16", asr_options=None, language=None,
vad_options=None, model=None, task="transcribe"):
=======
from .types import TranscriptionResult, SingleSegment
def load_model(whisper_arch, def load_model(whisper_arch,
device, device,
device_index=0, device_index=0,
@ -37,7 +31,6 @@ def load_model(whisper_arch,
model=None, model=None,
task="transcribe", task="transcribe",
download_root=None): download_root=None):
>>>>>>> ec6a110cdf2616919cfd0a616f9ae2fbdd44903f
'''Load a Whisper model for inference. '''Load a Whisper model for inference.
Args: Args:
whisper_arch: str - The name of the Whisper model to load. whisper_arch: str - The name of the Whisper model to load.
@ -100,9 +93,6 @@ def load_model(whisper_arch,
default_asr_options = faster_whisper.transcribe.TranscriptionOptions(**default_asr_options) default_asr_options = faster_whisper.transcribe.TranscriptionOptions(**default_asr_options)
default_vad_options = { default_vad_options = {
"vad_onset": 0.500, "vad_onset": 0.500,
"vad_offset": 0.363 "vad_offset": 0.363