-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.py
70 lines (56 loc) · 2.1 KB
/
app.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
from fastapi import FastAPI, HTTPException, UploadFile, File, Query
from fastapi import Response
from fastapi.responses import PlainTextResponse, JSONResponse
from fastapi.openapi.utils import get_openapi
import os
import requests
"""
API to Transcribe audio or video from URL
Generate Transcript in SRT, VTT, JSON and RAW format
"""
app = FastAPI()
TMP_DIR = "tmp/"
def fetch_audio_from_url(url: str) -> bytes:
response = requests.get(url)
response.raise_for_status()
return response.content
def transcribe_audio(audio_content: bytes, subtitle_format: str):
# Save the audio content to a temporary file
audio_file_path = os.path.join(TMP_DIR, "temp_audio.mp3")
with open(audio_file_path, "wb") as audio_file:
audio_file.write(audio_content)
# Use autosub to transcribe the original audio file
os.system(f"autosub {audio_file_path} --format {subtitle_format}")
# Read the generated subtitle file
subtitle_path = audio_file_path.replace(".mp3", f".{subtitle_format}")
with open(subtitle_path, 'r') as subtitle_file:
transcript = subtitle_file.read()
# Clean up temporary files
os.remove(subtitle_path)
os.remove(audio_file_path)
return transcript
@app.post("/transcribe")
async def transcribe(
url: str,
subtitle_format: str = Query("vtt", regex="^(srt|vtt|json|raw)$")
):
audio_content = fetch_audio_from_url(url)
if not audio_content:
raise HTTPException(status_code=500, \
detail="Failed to fetch audio from the provided URL.")
transcript = transcribe_audio(audio_content, subtitle_format)
if subtitle_format == 'json':
return Response(transcript, media_type="application/json")
return PlainTextResponse(transcript)
def custom_openapi():
if app.openapi_schema:
return app.openapi_schema
openapi_schema = get_openapi(
title="Speech To Text",
version="v1",
description="Generate Transcript in SRT, VTT, RAW and JSON format",
routes=app.routes
)
app.openapi_schema = openapi_schema
return app.openapi_schema
app.openapi = custom_openapi