Tonic commited on
Commit
737d2c6
1 Parent(s): 989551b

Delete app.py

Browse files
Files changed (1) hide show
  1. app.py +0 -82
app.py DELETED
@@ -1,82 +0,0 @@
1
- import gradio as gr
2
- import torchaudio
3
- import torch
4
- import os
5
- import time
6
- import soundfile as sf
7
-
8
- welcome_message = """
9
- # 👋🏻Welcome to 🔥Tonic's🗣️Unity On Device!🚀
10
-
11
- 🔥Tonic's🗣️Unity On Device!🚀 uses [facebook/seamless-m4t-unity-small](https://huggingface.co/facebook/seamless-m4t-unity-small) for audio translation & accessibility.
12
- 🔥Tonic's🗣️Unity On Device!🚀 on your own data & in your own way by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/TeamTonic/SeamlessOnDevice?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3>
13
- ### Join us :
14
- 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/GWpVpekp) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)"
15
- """
16
-
17
- # Define the list of target languages
18
- languages = {
19
- "English": "eng",
20
- "Hindi": "hin",
21
- "Portuguese": "por",
22
- "Russian": "rus",
23
- "Spanish": "spa"
24
- }
25
-
26
- def save_audio(audio_input, output_dir="saved_audio"):
27
- if not os.path.exists(output_dir):
28
- os.makedirs(output_dir)
29
-
30
- # Extract sample rate and audio data
31
- sample_rate, audio_data = audio_input
32
-
33
- # Generate a unique file name
34
- file_name = f"audio_{int(time.time())}.wav"
35
- file_path = os.path.join(output_dir, file_name)
36
-
37
- # Save the audio file
38
- sf.write(file_path, audio_data, sample_rate)
39
-
40
- return file_path
41
-
42
- def speech_to_text(audio_data, tgt_lang):
43
- file_path = save_audio(audio_data)
44
- audio_input, _ = torchaudio.load(file_path)
45
- s2t_model = torch.jit.load("unity_on_device_s2t.ptl")
46
- with torch.no_grad():
47
- text = s2t_model(audio_input, tgt_lang=languages[tgt_lang])
48
- return text
49
-
50
- def speech_to_speech_translation(audio_data, tgt_lang):
51
- file_path = save_audio(audio_data)
52
- audio_input, _ = torchaudio.load(file_path)
53
- s2st_model = torch.jit.load("unity_on_device.ptl")
54
- with torch.no_grad():
55
- text, units, waveform = s2st_model(audio_input, tgt_lang=languages[tgt_lang])
56
- output_file = "/tmp/result.wav"
57
- torchaudio.save(output_file, waveform.unsqueeze(0), sample_rate=16000)
58
- return text, output_file
59
-
60
- def create_interface():
61
- with gr.Markdown(welcome_message)
62
- with gr.Blocks(theme='ParityError/Anime') as interface:
63
- # Dropdown for language selection
64
- input_language = gr.Dropdown(list(languages.keys()), label="Select Target Language", value="English")
65
-
66
- with gr.Accordion("Speech to Text", open=False) as stt_accordion:
67
- audio_input_stt = gr.Audio(label="Upload or Record Audio")
68
- text_output_stt = gr.Text(label="Transcribed Text")
69
- stt_button = gr.Button("Transcribe")
70
- stt_button.click(speech_to_text, inputs=[audio_input_stt, input_language], outputs=text_output_stt)
71
-
72
- with gr.Accordion("Speech to Speech Translation", open=False) as s2st_accordion:
73
- audio_input_s2st = gr.Audio(label="Upload or Record Audio")
74
- text_output_s2st = gr.Text(label="Translated Text")
75
- audio_output_s2st = gr.Audio(label="Translated Audio", type="filepath")
76
- s2st_button = gr.Button("Translate")
77
- s2st_button.click(speech_to_speech_translation, inputs=[audio_input_s2st, input_language], outputs=[text_output_s2st, audio_output_s2st])
78
-
79
- return interface
80
-
81
- app = create_interface()
82
- app.launch(show_error=True, debug=True)