add fastapi usage

README.md

@@ -135,6 +135,8 @@ docker build -t cosyvoice:v1.0 .
 docker run -d --runtime=nvidia -p 50000:50000 cosyvoice:v1.0 /bin/bash -c "cd /opt/CosyVoice/CosyVoice/runtime/python/grpc && python3 server.py --port 50000 --max_conc 4 --model_dir iic/CosyVoice-300M && sleep infinity"
 python3 grpc/client.py --port 50000 --mode <sft|zero_shot|cross_lingual|instruct>
 # for fastapi usage
+docker run -d --runtime=nvidia -p 50000:50000 cosyvoice:v1.0 /bin/bash -c "cd /opt/CosyVoice/CosyVoice/runtime/python/fastapi && MODEL_DIR=iic/CosyVoice-300M fastapi dev --port 50000 server.py && sleep infinity"
+python3 fastapi/client.py --port 50000 --mode <sft|zero_shot|cross_lingual|instruct>
 ```
 
 ## Discussion & Communication

runtime/python/fastapi/fastapi_client.py → runtime/python/fastapi/client.py

@@ -66,7 +66,7 @@ if __name__ == "__main__":
                         default='希望你以后能够做的比我还好呦。')
     parser.add_argument('--prompt_wav',
                         type=str,
-                        default='../../zero_shot_prompt.wav')
+                        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.')

runtime/python/fastapi/fastapi_server.py → runtime/python/fastapi/server.py

@@ -1,9 +1,9 @@
 # Set inference model
 # export MODEL_DIR=pretrained_models/CosyVoice-300M-Instruct
 # For development
-# fastapi dev --port 6006 fastapi_server.py 
+# fastapi dev --port 6006 fastapi_server.py
 # For production deployment
-# fastapi run --port 6006 fastapi_server.py 
+# fastapi run --port 6006 fastapi_server.py
 
 import os
 import sys
@@ -12,8 +12,8 @@ from fastapi import FastAPI, Response, File, UploadFile, Form
 from fastapi.responses import HTMLResponse
 from contextlib import asynccontextmanager
 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))
+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
 import numpy as np
@@ -30,7 +30,7 @@ async def lifespan(app: FastAPI):
     model_dir = os.getenv("MODEL_DIR", "pretrained_models/CosyVoice-300M-SFT")
     if model_dir:
         logging.info("MODEL_DIR is {}", model_dir)
-        app.cosyvoice = CosyVoice('../../'+model_dir)
+        app.cosyvoice = CosyVoice(model_dir)
         # sft usage
         logging.info("Avaliable speakers {}", app.cosyvoice.list_avaliable_spks())
     else: