deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 18:09:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix lint

Browse Source
This commit is contained in:
lyuxiang.lx
2024-09-05 16:15:34 +08:00
parent eeebc45313
commit 90433f5373
35 changed files with 189 additions and 122 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
runtime/python/fastapi/client.py
View File

@@ -38,7 +38,7 @@ def main():
payload = {
'tts_text': args.tts_text,
}
files = [('prompt_wav', ('prompt_wav', open(args.prompt_wav,'rb'), 'application/octet-stream'))]
files = [('prompt_wav', ('prompt_wav', open(args.prompt_wav, 'rb'), 'application/octet-stream'))]
response = requests.request("GET", url, data=payload, files=files, stream=True)
else:
payload = {
@@ -55,6 +55,7 @@ def main():
torchaudio.save(args.tts_wav, tts_speech, target_sr)
logging.info('get response')
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('--host',
@@ -81,7 +82,8 @@ if __name__ == "__main__":
default='../../../zero_shot_prompt.wav')
parser.add_argument('--instruct_text',
type=str,
default='Theo \'Crimson\', is a fiery, passionate rebel leader. Fights with fervor for justice, but struggles with impulsiveness.')
default='Theo \'Crimson\', is a fiery, passionate rebel leader. \
Fights with fervor for justice, but struggles with impulsiveness.')
parser.add_argument('--tts_wav',
type=str,
default='demo.wav')

16
runtime/python/fastapi/server.py
View File

@@ -13,9 +13,6 @@
# limitations under the License.
import os
import sys
ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
sys.path.append('{}/../../..'.format(ROOT_DIR))
sys.path.append('{}/../../../third_party/Matcha-TTS'.format(ROOT_DIR))
import argparse
import logging
logging.getLogger('matplotlib').setLevel(logging.WARNING)
@@ -24,6 +21,9 @@ from fastapi.responses import StreamingResponse
from fastapi.middleware.cors import CORSMiddleware
import uvicorn
import numpy as np
ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
sys.path.append('{}/../../..'.format(ROOT_DIR))
sys.path.append('{}/../../../third_party/Matcha-TTS'.format(ROOT_DIR))
from cosyvoice.cli.cosyvoice import CosyVoice
from cosyvoice.utils.file_utils import load_wav
@@ -36,34 +36,40 @@ app.add_middleware(
allow_methods=["*"],
allow_headers=["*"])
def generate_data(model_output):
for i in model_output:
tts_audio = (i['tts_speech'].numpy() * (2 ** 15)).astype(np.int16).tobytes()
yield tts_audio
@app.get("/inference_sft")
async def inference_sft(tts_text: str = Form(), spk_id: str = Form()):
model_output = cosyvoice.inference_sft(tts_text, spk_id)
return StreamingResponse(generate_data(model_output))
@app.get("/inference_zero_shot")
async def inference_zero_shot(tts_text: str = Form(), prompt_text: str = Form(), prompt_wav: UploadFile = File()):
prompt_speech_16k = load_wav(prompt_wav.file, 16000)
model_output = cosyvoice.inference_zero_shot(tts_text, prompt_text, prompt_speech_16k)
return StreamingResponse(generate_data(model_output))
@app.get("/inference_cross_lingual")
async def inference_cross_lingual(tts_text: str = Form(), prompt_wav: UploadFile = File()):
prompt_speech_16k = load_wav(prompt_wav.file, 16000)
model_output = cosyvoice.inference_cross_lingual(tts_text, prompt_speech_16k)
return StreamingResponse(generate_data(model_output))
@app.get("/inference_instruct")
async def inference_instruct(tts_text: str = Form(), spk_id: str = Form(), instruct_text: str = Form()):
model_output = cosyvoice.inference_instruct(tts_text, spk_id, instruct_text)
return StreamingResponse(generate_data(model_output))
if __name__=='__main__':
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--port',
type=int,
@@ -74,4 +80,4 @@ if __name__=='__main__':
help='local path or modelscope repo id')
args = parser.parse_args()
cosyvoice = CosyVoice(args.model_dir)
uvicorn.run(app, host="127.0.0.1", port=args.port)
uvicorn.run(app, host="127.0.0.1", port=args.port)