You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
80 lines
2.3 KiB
80 lines
2.3 KiB
1 year ago
|
import os
|
||
|
|
|
||
|
|
DEBUG = os.environ.get('DEBUG')
|
||
|
|
|
||
|
|
import logging
|
||
|
|
logging.basicConfig(
|
||
|
|
format='[%(asctime)s] %(levelname)s %(module)s/%(funcName)s: - %(message)s',
|
||
|
|
level=logging.DEBUG if DEBUG else logging.INFO)
|
||
|
|
logging.getLogger('aiohttp').setLevel(logging.DEBUG if DEBUG else logging.WARNING)
|
||
|
|
|
||
|
|
|
||
|
|
import ffmpeg
|
||
|
|
import whisper
|
||
|
|
import time
|
||
|
|
import asyncio
|
||
|
|
from aiohttp import web, ClientSession, ClientError
|
||
|
|
import numpy as np
|
||
|
|
|
||
|
|
app = web.Application()
|
||
|
|
PORT = 3002
|
||
|
|
SAMPLE_RATE = 16000
|
||
|
|
|
||
|
|
logging.info('Loading whisper model...')
|
||
|
|
model = whisper.load_model('medium')
|
||
|
|
logging.info('Done.')
|
||
|
|
|
||
|
|
#start = time.time()
|
||
|
|
#result = model.transcribe('whisper-test.ogg')
|
||
|
|
#print('finished in', time.time() - start, 's')
|
||
|
|
#
|
||
|
|
#print(result['text'])
|
||
|
|
|
||
|
|
def load_audio(binary_file, sr = SAMPLE_RATE):
|
||
|
|
# stolen from https://github.com/ckaytev/tgisper/blob/main/tgisper/tgisperbot.py
|
||
|
|
try:
|
||
|
|
# This launches a subprocess to decode audio while down-mixing and
|
||
|
|
# resampling as necessary.
|
||
|
|
# Requires the ffmpeg CLI and `ffmpeg-python` package to be installed.
|
||
|
|
out, _ = (
|
||
|
|
ffmpeg.input("pipe:", threads=0)
|
||
|
|
.output("-", format="s16le", acodec="pcm_s16le", ac=1, ar=sr)
|
||
|
|
.run(cmd="ffmpeg", capture_stdout=True, capture_stderr=True, input=binary_file)
|
||
|
|
)
|
||
|
|
except ffmpeg.Error as e:
|
||
|
|
raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
|
||
|
|
|
||
|
|
return np.frombuffer(out, np.int16).flatten().astype(np.float32) / 32768.0
|
||
|
|
|
||
|
|
async def index(request):
|
||
|
|
return web.Response(text='hello world', content_type='text/html')
|
||
|
|
|
||
|
|
async def post_whisper(request):
|
||
|
|
data = await request.post()
|
||
|
|
audio = load_audio(data['audio'].file.read())
|
||
|
|
|
||
|
|
logging.info('Starting audio transcription...')
|
||
|
|
result = model.transcribe(audio)
|
||
|
|
logging.info('Done.')
|
||
|
|
|
||
|
|
return web.json_response(result)
|
||
|
|
|
||
|
|
async def run_webserver():
|
||
|
|
logging.info('Starting webserver on port: %s', PORT)
|
||
|
|
runner = web.AppRunner(app)
|
||
|
|
await runner.setup()
|
||
|
|
site = web.TCPSite(runner, '0.0.0.0', PORT)
|
||
|
|
await site.start()
|
||
|
|
|
||
|
|
while True:
|
||
|
|
await asyncio.sleep(10)
|
||
|
|
|
||
|
|
if __name__ == '__main__':
|
||
|
|
app.router.add_get('/', index)
|
||
|
|
app.router.add_post('/whisper', post_whisper)
|
||
|
|
|
||
|
|
loop = asyncio.get_event_loop()
|
||
|
|
a = loop.create_task(run_webserver())
|
||
|
|
loop.run_forever()
|
||
|
|
|