open-webui/backend/apps/audio/main.py

214 lines
5.5 KiB
Python
Raw Permalink Normal View History

import os
import logging
2024-02-11 09:17:50 +01:00
from fastapi import (
FastAPI,
Request,
Depends,
HTTPException,
status,
UploadFile,
File,
Form,
)
2024-04-20 22:15:59 +02:00
from fastapi.responses import StreamingResponse, JSONResponse, FileResponse
2024-02-11 09:17:50 +01:00
from fastapi.middleware.cors import CORSMiddleware
from faster_whisper import WhisperModel
2024-04-20 22:21:52 +02:00
from pydantic import BaseModel
2024-02-11 09:17:50 +01:00
2024-04-20 22:15:59 +02:00
import requests
import hashlib
from pathlib import Path
import json
2024-02-11 09:17:50 +01:00
from constants import ERROR_MESSAGES
from utils.utils import (
decode_token,
get_current_user,
get_verified_user,
get_admin_user,
)
from utils.misc import calculate_sha256
2024-03-31 10:13:39 +02:00
from config import (
SRC_LOG_LEVELS,
CACHE_DIR,
UPLOAD_DIR,
WHISPER_MODEL,
WHISPER_MODEL_DIR,
WHISPER_MODEL_AUTO_UPDATE,
2024-04-02 14:47:52 +02:00
DEVICE_TYPE,
2024-04-20 23:04:16 +02:00
AUDIO_OPENAI_API_BASE_URL,
AUDIO_OPENAI_API_KEY,
2024-03-31 10:13:39 +02:00
)
log = logging.getLogger(__name__)
log.setLevel(SRC_LOG_LEVELS["AUDIO"])
2024-02-11 09:17:50 +01:00
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
2024-04-20 22:15:59 +02:00
2024-04-20 23:04:16 +02:00
app.state.OPENAI_API_BASE_URL = AUDIO_OPENAI_API_BASE_URL
app.state.OPENAI_API_KEY = AUDIO_OPENAI_API_KEY
2024-04-20 22:15:59 +02:00
2024-04-02 14:47:52 +02:00
# setting device type for whisper model
whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu"
log.info(f"whisper_device_type: {whisper_device_type}")
2024-04-20 22:15:59 +02:00
SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
2024-04-20 22:21:52 +02:00
class OpenAIConfigUpdateForm(BaseModel):
url: str
key: str
@app.get("/config")
async def get_openai_config(user=Depends(get_admin_user)):
return {
"OPENAI_API_BASE_URL": app.state.OPENAI_API_BASE_URL,
"OPENAI_API_KEY": app.state.OPENAI_API_KEY,
}
@app.post("/config/update")
async def update_openai_config(
form_data: OpenAIConfigUpdateForm, user=Depends(get_admin_user)
):
if form_data.key == "":
raise HTTPException(status_code=400, detail=ERROR_MESSAGES.API_KEY_NOT_FOUND)
app.state.OPENAI_API_BASE_URL = form_data.url
app.state.OPENAI_API_KEY = form_data.key
return {
"status": True,
"OPENAI_API_BASE_URL": app.state.OPENAI_API_BASE_URL,
"OPENAI_API_KEY": app.state.OPENAI_API_KEY,
}
2024-04-20 22:15:59 +02:00
@app.post("/speech")
async def speech(request: Request, user=Depends(get_verified_user)):
2024-04-20 23:00:24 +02:00
body = await request.body()
name = hashlib.sha256(body).hexdigest()
file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
# Check if the file already exists in the cache
if file_path.is_file():
return FileResponse(file_path)
headers = {}
headers["Authorization"] = f"Bearer {app.state.OPENAI_API_KEY}"
headers["Content-Type"] = "application/json"
r = None
2024-04-20 22:15:59 +02:00
try:
2024-04-20 23:00:24 +02:00
r = requests.post(
url=f"{app.state.OPENAI_API_BASE_URL}/audio/speech",
data=body,
headers=headers,
stream=True,
)
r.raise_for_status()
# Save the streaming content to a file
with open(file_path, "wb") as f:
for chunk in r.iter_content(chunk_size=8192):
f.write(chunk)
with open(file_body_path, "w") as f:
json.dump(json.loads(body.decode("utf-8")), f)
# Return the saved file
return FileResponse(file_path)
except Exception as e:
log.exception(e)
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"External: {res['error']['message']}"
except:
error_detail = f"External: {e}"
raise HTTPException(
status_code=r.status_code if r != None else 500,
detail=error_detail,
)
2024-04-20 22:15:59 +02:00
2024-02-11 09:17:50 +01:00
2024-04-20 22:15:59 +02:00
@app.post("/transcriptions")
2024-02-11 09:17:50 +01:00
def transcribe(
file: UploadFile = File(...),
user=Depends(get_current_user),
):
log.info(f"file.content_type: {file.content_type}")
2024-02-11 09:17:50 +01:00
if file.content_type not in ["audio/mpeg", "audio/wav"]:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED,
)
try:
filename = file.filename
file_path = f"{UPLOAD_DIR}/{filename}"
contents = file.file.read()
with open(file_path, "wb") as f:
f.write(contents)
f.close()
whisper_kwargs = {
"model_size_or_path": WHISPER_MODEL,
"device": whisper_device_type,
"compute_type": "int8",
"download_root": WHISPER_MODEL_DIR,
"local_files_only": not WHISPER_MODEL_AUTO_UPDATE,
}
log.debug(f"whisper_kwargs: {whisper_kwargs}")
2024-04-11 04:44:44 +02:00
try:
model = WhisperModel(**whisper_kwargs)
except:
2024-04-13 11:18:13 +02:00
log.warning(
2024-04-11 04:44:44 +02:00
"WhisperModel initialization failed, attempting download with local_files_only=False"
)
whisper_kwargs["local_files_only"] = False
model = WhisperModel(**whisper_kwargs)
2024-02-11 09:17:50 +01:00
segments, info = model.transcribe(file_path, beam_size=5)
log.info(
2024-02-11 09:17:50 +01:00
"Detected language '%s' with probability %f"
% (info.language, info.language_probability)
)
transcript = "".join([segment.text for segment in list(segments)])
2024-02-11 11:17:24 +01:00
return {"text": transcript.strip()}
2024-02-11 09:17:50 +01:00
except Exception as e:
log.exception(e)
2024-02-11 09:17:50 +01:00
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=ERROR_MESSAGES.DEFAULT(e),
)