Spaces:
Sleeping
Sleeping
v3
Browse files
app.py
CHANGED
@@ -2,6 +2,26 @@ import gradio as gr
|
|
2 |
import whisper
|
3 |
from pytube import YouTube
|
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
def get_audio(url):
|
6 |
print(f'{url} start get audio ...')
|
7 |
yt = YouTube(url)
|
@@ -10,6 +30,20 @@ def get_audio(url):
|
|
10 |
return audio_file
|
11 |
|
12 |
def get_transcript(url, model_size, lang, format):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
|
14 |
|
15 |
print('whisper loading ...')
|
|
|
2 |
import whisper
|
3 |
from pytube import YouTube
|
4 |
|
5 |
+
|
6 |
+
import requests
|
7 |
+
|
8 |
+
headers = {
|
9 |
+
'accept': 'application/json',
|
10 |
+
'x-gladia-key': '89b0adf5-fb2c-48ba-8a66-76b02827fd14',
|
11 |
+
# requests won't add a boundary if this header is set when you pass files=
|
12 |
+
# 'Content-Type': 'multipart/form-data',
|
13 |
+
}
|
14 |
+
|
15 |
+
files = {
|
16 |
+
'audio': ("707539ca80d090a28c5ea7bbf93e8068.mp4", open('707539ca80d090a28c5ea7bbf93e8068.mp4', 'rb'), 'video/mp4'),
|
17 |
+
'audio_url': (None, 'http://files.gladia.io/example/audio-transcription/split_infinity.wav'),
|
18 |
+
'language': (None, 'english'),
|
19 |
+
'language_behaviour': (None, 'automatic single language'),
|
20 |
+
}
|
21 |
+
|
22 |
+
response = requests.post('https://api.gladia.io/audio/text/audio-transcription/', headers=headers, files=files)
|
23 |
+
|
24 |
+
|
25 |
def get_audio(url):
|
26 |
print(f'{url} start get audio ...')
|
27 |
yt = YouTube(url)
|
|
|
30 |
return audio_file
|
31 |
|
32 |
def get_transcript(url, model_size, lang, format):
|
33 |
+
audio_file = get_audio(url)
|
34 |
+
audio_file = 'tmp.mp4'
|
35 |
+
files = {
|
36 |
+
'audio': (f"{audio_file}", open(f'{audio_file}', 'rb'), 'video/mp4'),
|
37 |
+
'audio_url': (None, 'http://files.gladia.io/example/audio-transcription/split_infinity.wav'),
|
38 |
+
'language': (None, 'english'),
|
39 |
+
'language_behaviour': (None, 'automatic single language'),
|
40 |
+
}
|
41 |
+
response = requests.post('https://api.gladia.io/audio/text/audio-transcription/', headers=headers, files=files)
|
42 |
+
return response.text
|
43 |
+
|
44 |
+
|
45 |
+
|
46 |
+
def get_transcript2(url, model_size, lang, format):
|
47 |
|
48 |
|
49 |
print('whisper loading ...')
|