artificialguybr commited on
Commit
bea60dd
β€’
1 Parent(s): c480a67

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +75 -0
app.py ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+ import subprocess
3
+ import whisper
4
+ from googletrans import Translator
5
+ import asyncio
6
+ import edge_tts
7
+ import os
8
+
9
+ # Extract and Transcribe Audio
10
+ def extract_and_transcribe_audio(video_path):
11
+ ffmpeg_command = f"ffmpeg -i '{video_path}' -acodec pcm_s24le -ar 48000 -q:a 0 -map a -y 'output_audio.wav'"
12
+ subprocess.run(ffmpeg_command, shell=True)
13
+ model = whisper.load_model("base")
14
+ result = model.transcribe("output_audio.wav")
15
+ return result["text"], result['language']
16
+
17
+ # Translate Text
18
+ def translate_text(whisper_text, whisper_language, target_language):
19
+ language_mapping = {
20
+ 'English': 'en',
21
+ 'Spanish': 'es',
22
+ # ... (other mappings)
23
+ }
24
+ target_language_code = language_mapping[target_language]
25
+ translator = Translator()
26
+ translated_text = translator.translate(whisper_text, src=whisper_language, dest=target_language_code).text
27
+ return translated_text
28
+
29
+ # Generate Voice
30
+ async def generate_voice(translated_text, target_language):
31
+ VOICE_MAPPING = {
32
+ 'English': 'en-GB-SoniaNeural',
33
+ 'Spanish': 'es-ES-PabloNeural',
34
+ # ... (other mappings)
35
+ }
36
+ voice = VOICE_MAPPING[target_language]
37
+ communicate = edge_tts.Communicate(translated_text, voice)
38
+ await communicate.save("output_synth.wav")
39
+ return "output_synth.wav"
40
+
41
+ # Generate Lip-synced Video (Placeholder)
42
+ def generate_lip_synced_video(video_path, output_audio_path):
43
+ # Your lip-synced video generation code here
44
+ # ...
45
+ return "output_high_qual.mp4"
46
+
47
+ # Main function to be called by Gradio
48
+ def process_video(video, target_language):
49
+ video_path = "uploaded_video.mp4"
50
+ with open(video_path, "wb") as f:
51
+ f.write(video.read())
52
+
53
+ # Step 1: Extract and Transcribe Audio
54
+ whisper_text, whisper_language = extract_and_transcribe_audio(video_path)
55
+
56
+ # Step 2: Translate Text
57
+ translated_text = translate_text(whisper_text, whisper_language, target_language)
58
+
59
+ # Step 3: Generate Voice
60
+ loop = asyncio.get_event_loop()
61
+ output_audio_path = loop.run_until_complete(generate_voice(translated_text, target_language))
62
+
63
+ # Step 4: Generate Lip-synced Video
64
+ output_video_path = generate_lip_synced_video(video_path, output_audio_path)
65
+
66
+ return output_video_path
67
+
68
+ # Gradio Interface
69
+ iface = gr.Interface(
70
+ fn=process_video,
71
+ inputs=["file", gr.Interface.Component(type="dropdown", choices=["English", "Spanish"])],
72
+ outputs="file",
73
+ live=False
74
+ )
75
+ iface.launch()