speech_recognition/SenseVoice
1
0
Fork 0
mirror of https://github.com/FunAudioLLM/SenseVoice.git synced 2025-09-15 15:08:35 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #75 from godxiaoji/fix-api

Browse Source 
fix api torchaudio.load bug
This commit is contained in:
zhifu gao 2024-07-29 09:58:48 +08:00 committed by GitHub
commit 797a9a5f9c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
api.py
View File

@ -10,6 +10,8 @@ from enum import Enum
import torchaudio
from model import SenseVoiceSmall
from funasr.utils.postprocess_utils import rich_transcription_postprocess
from io import BytesIO
class Language(str, Enum):
auto = "auto"
@ -49,9 +51,11 @@ async def turn_audio_to_text(files: Annotated[List[bytes], File(description="wav
audios = []
audio_fs = 0
for file in files:
data_or_path_or_list, audio_fs = torchaudio.load(file)
file_io = BytesIO(file)
data_or_path_or_list, audio_fs = torchaudio.load(file_io)
data_or_path_or_list = data_or_path_or_list.mean(0)
audios.append(data_or_path_or_list)
file_io.close()
if lang == "":
lang = "auto"
if keys == "":