File size: 2,035 Bytes
d7d0d8e 252d749 4fceeff 88c5489 252d749 385a3d3 252d749 d7d0d8e 252d749 647122b 9d130ca 647122b 252d749 d7d0d8e 88c5489 d7d0d8e 4fceeff 136640b 4fceeff d7d0d8e 385a3d3 78f1df3 7a1123a d7d0d8e 385a3d3 af60251 385a3d3 7306982 dd2695b af5c58a 7306982 385a3d3 dd2695b 7a1123a 88c5489 d7d0d8e dd2695b d7d0d8e |
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 |
from celery import Celery, chain
import os
import time,subprocess
import cgi
from App import celery_config
import yt_dlp
import tempfile
from App.Transcription.Utils.audio_transcription import transcribe_file
from App.Embedding.utils.Initialize import encode, generateChunks
celery = Celery()
celery.config_from_object(celery_config)
celery.conf.update(
# Other Celery configuration settings
CELERYD_LOG_LEVEL='INFO', # Set log level to DEBUG for the worker
)
@celery.task(name="embbeding", bind=True)
def generate_store(self, data, task_id):
print('generating chunks')
chunks = generateChunks(data, task_id)
encode(chunks)
print("hellooo")
def download_with_wget(link, download_dir, filename):
subprocess.run(["aria2c", link, "-d", download_dir, "-o", filename])
@celery.task(name="download", bind=True)
def downloadUrl(self, link, download_dir, filename, model_size="base"):
file_path=os.path.join(download_dir,filename)
download_with_wget(link=link,download_dir=download_dir,filename=filename)
data = transcribe_file(state=self, file_path=file_path, model_size=model_size)
# generate_store.delay(data["content"], self.request.id)
return data
@celery.task(name="transcription", bind=True)
def transcription_task(self, file_path, model_size="tiny"):
data = transcribe_file(state=self, file_path=file_path, model_size=model_size)
generate_store.delay(data["content"], self.request.id)
return data
@celery.task(name="download", bind=True)
def downloadfile(self, url, ydl_opts, model_size="base"):
self.update_state(state="Downloading File..", meta={})
with yt_dlp.YoutubeDL(ydl_opts) as ydl:
ydl.download([url])
# updated
self.update_state(state="Downloading complete", meta={})
audio_file = ydl_opts["outtmpl"]
print(model_size, "worker after")
data = transcribe_file(
state=self, file_path=audio_file["default"], model_size=model_size
)
generate_store.delay(data["content"], self.request.id)
return data
|