rixty
/
VoiceCraft
mirror of https://github.com/jasonppy/VoiceCraft.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #140 from Approximetal/patch-1 

Remove extra whitespaces to prevent unwanted intonation
This commit is contained in:
Puyuan Peng 2024-06-24 10:25:51 -04:00 committed by GitHub
parent 013a21c70d 82d5676019
commit 4873249ba3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
gradio_app.py
View File

@ -21,7 +21,7 @@ TMP_PATH = os.getenv("TMP_PATH", "./demo/temp")
MODELS_PATH = os.getenv("MODELS_PATH", "./pretrained_models") MODELS_PATH = os.getenv("MODELS_PATH", "./pretrained_models")
device = "cuda" if torch.cuda.is_available() else "cpu" device = "cuda" if torch.cuda.is_available() else "cpu"
whisper_model, align_model, voicecraft_model = None, None, None whisper_model, align_model, voicecraft_model = None, None, None
_whitespace_re = re.compile(r"\s+")
def get_random_string(): def get_random_string():
return "".join(str(uuid.uuid4()).split("-")) return "".join(str(uuid.uuid4()).split("-"))
@ -270,7 +270,7 @@ def run(seed, left_margin, right_margin, codec_audio_sr, codec_sr, top_k, top_p,
target_transcript = sentence target_transcript = sentence
inference_transcript += target_transcript + "\n" inference_transcript += target_transcript + "\n"
target_transcript = re.sub(_whitespace_re, " ", target_transcript)
prompt_end_frame = int(min(audio_dur, prompt_end_time) * info.sample_rate) prompt_end_frame = int(min(audio_dur, prompt_end_time) * info.sample_rate)
_, gen_audio = inference_one_sample(voicecraft_model["model"], _, gen_audio = inference_one_sample(voicecraft_model["model"],
voicecraft_model["config"], voicecraft_model["config"],
@ -296,7 +296,7 @@ def run(seed, left_margin, right_margin, codec_audio_sr, codec_sr, top_k, top_p,
target_transcript = sentence target_transcript = sentence
inference_transcript += target_transcript + "\n" inference_transcript += target_transcript + "\n"
target_transcript = re.sub(_whitespace_re, " ", target_transcript)
morphed_span = (max(edit_start_time - left_margin, 1 / codec_sr), min(edit_end_time + right_margin, audio_dur)) morphed_span = (max(edit_start_time - left_margin, 1 / codec_sr), min(edit_end_time + right_margin, audio_dur))
mask_interval = [[round(morphed_span[0]*codec_sr), round(morphed_span[1]*codec_sr)]] mask_interval = [[round(morphed_span[0]*codec_sr), round(morphed_span[1]*codec_sr)]]
mask_interval = torch.LongTensor(mask_interval) mask_interval = torch.LongTensor(mask_interval)