Spaces:
Sleeping
Sleeping
Commit
·
21621d2
1
Parent(s):
2aea2e8
Update app.py
Browse files
app.py
CHANGED
|
@@ -6,20 +6,29 @@ import transformers
|
|
| 6 |
from transformers import pipeline
|
| 7 |
import torch
|
| 8 |
from summarizer import Summarizer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
#Download youtube video
|
| 11 |
def download_audio(link):
|
| 12 |
with yt_dlp.YoutubeDL({'extract_audio': True, 'format': 'bestaudio', 'outtmpl': 'video.mp3'}) as video:
|
| 13 |
-
info_dict = video.extract_info(link, download = True)
|
| 14 |
-
video_title = info_dict['title']
|
| 15 |
video.download(link)
|
| 16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
#def portuguese_sum_pipeline(link):
|
| 19 |
# video_title = download_audio(link)
|
| 20 |
|
| 21 |
def english_sum_pipeline(link):
|
| 22 |
-
|
| 23 |
|
| 24 |
#audio-to-text
|
| 25 |
transcriptor = whisper.load_model("base.en")
|
|
|
|
| 6 |
from transformers import pipeline
|
| 7 |
import torch
|
| 8 |
from summarizer import Summarizer
|
| 9 |
+
import ffmpeg
|
| 10 |
+
import speech_recognition as sr
|
| 11 |
+
from pytube import YouTube
|
| 12 |
+
import pandas as pd
|
| 13 |
+
import string
|
| 14 |
+
import whisper
|
| 15 |
+
from moviepy.editor import AudioFileClip
|
| 16 |
|
| 17 |
#Download youtube video
|
| 18 |
def download_audio(link):
|
| 19 |
with yt_dlp.YoutubeDL({'extract_audio': True, 'format': 'bestaudio', 'outtmpl': 'video.mp3'}) as video:
|
|
|
|
|
|
|
| 20 |
video.download(link)
|
| 21 |
+
|
| 22 |
+
#Convert video format
|
| 23 |
+
def convert_mp4_to_wav(mp4_file, wav_file):
|
| 24 |
+
video = AudioFileClip(mp4_file)
|
| 25 |
+
video.write_audiofile(wav_file)
|
| 26 |
|
| 27 |
#def portuguese_sum_pipeline(link):
|
| 28 |
# video_title = download_audio(link)
|
| 29 |
|
| 30 |
def english_sum_pipeline(link):
|
| 31 |
+
download_audio(link)
|
| 32 |
|
| 33 |
#audio-to-text
|
| 34 |
transcriptor = whisper.load_model("base.en")
|