Hugging Face's logo

Spaces:
YashMachineLearning
/
indicF5
Sleeping

App Files Community
ashishkblink commited on
Commit
72ab360
·
verified ·
1 Parent(s): fb90de7

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +49 -38
README.md CHANGED
@@ -1,61 +1,72 @@
1
  ---
2
- title: Vakya TTS Playground
3
- emoji: 🎤
4
- colorFrom: purple
5
- colorTo: pink
6
  sdk: gradio
7
- sdk_version: 6.2.0
8
  app_file: app.py
9
  pinned: false
10
- license: apache-2.0
11
  ---
12
 
13
- # 🎤 Vakya TTS Playground
14
 
15
- **India's No. 1 TTS Model for Hindi and Other Indian Languages**
16
 
17
- Interactive playground to test and experience the power of Vakya TTS - a state-of-the-art Text-to-Speech model fine-tuned from XTTS-v2, specifically optimized for Hindi and other Indian languages.
18
 
19
- ## 🎯 Features
20
-
21
- - **High-quality Hindi TTS** - Optimized specifically for Hindi pronunciation and intonation
22
- - **Multi-Indian Language Support** - Supports 10+ Indian languages
23
- - **Voice Cloning** - Clone voices from just 6 seconds of audio
24
- - **Real-time Synthesis** - Fast and efficient speech generation
25
- - **Natural Sounding** - Human-like voice quality
26
 
27
  ## 🚀 How to Use
28
 
29
- 1. **Enter Text**: Type or paste your text in the text box
30
- 2. **Select Language**: Choose from Hindi, English, Marathi, Telugu, Tamil, Kannada, Gujarati, Punjabi, Bengali, or Urdu
31
- 3. **Upload Speaker Audio (Optional)**: Upload a 6+ second audio file to clone the voice
32
- 4. **Generate**: Click "Generate Speech" and enjoy the output!
 
 
 
 
 
 
33
 
34
- ## 📊 Supported Languages
 
 
 
 
 
 
35
 
36
- - Hindi (hi) - Primary focus
37
- - English (en)
38
- - Marathi (mr)
39
- - Telugu (te)
40
- - Tamil (ta)
41
- - Kannada (kn)
42
- - Gujarati (gu)
43
- - Punjabi (pa)
44
- - Bengali (bn)
45
- - Urdu (ur)
46
 
47
- ## 🔗 Model Repository
 
 
 
48
 
49
- The model is available at: [ashishkblink/vakya](https://huggingface.co/ashishkblink/vakya)
50
 
51
- ## 📄 License
 
 
 
52
 
53
- Apache 2.0
54
 
55
- ## 👤 Author
 
 
56
 
57
- ashishkblink
 
 
58
 
59
  ---
60
 
61
- *Built with ❤️ for the Indian language community*
 
 
1
  ---
2
+ title: Vakya 2.0 - Text-to-Speech
3
+ emoji: 🎙️
4
+ colorFrom: blue
5
+ colorTo: purple
6
  sdk: gradio
7
+ sdk_version: 4.0.0
8
  app_file: app.py
9
  pinned: false
10
+ license: mit
11
  ---
12
 
13
+ # 🎙️ Vakya 2.0 - Text-to-Speech Playground
14
 
15
+ **Vakya** is a high-quality Text-to-Speech model based on the IndicF5 architecture, supporting **11 Indian languages**.
16
 
17
+ ## 🌟 Features
18
 
19
+ - **Multi-language Support**: Assamese, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, Telugu
20
+ - **Voice Cloning**: Uses reference audio to clone voice characteristics
21
+ - **High Quality**: 24kHz sample rate, 0.4B parameter model
22
+ - **Easy to Use**: Simple interface for testing and experimentation
 
 
 
23
 
24
  ## 🚀 How to Use
25
 
26
+ 1. **Load Model**: Click the "Load Model" button (first time may take a few minutes to download)
27
+ 2. **Upload Reference Audio**: Upload a short audio clip (<15 seconds recommended) that represents the voice you want to clone
28
+ 3. **Enter Reference Text** (Optional): Type what is spoken in the reference audio. If left blank, the model will auto-transcribe it
29
+ 4. **Enter Text to Generate**: Type the text you want to synthesize in any supported language
30
+ 5. **Adjust Settings** (Optional):
31
+ - Speed: Control the speech rate (0.5x to 2.0x)
32
+ - Remove Silences: Experimental feature to remove pauses
33
+ 6. **Generate**: Click "Generate Speech" and wait for the audio output
34
+
35
+ ## 📋 Model Information
36
 
37
+ - **Model**: Vakya 2.0
38
+ - **Repository**: [ashishkblink/vakya2.0](https://huggingface.co/ashishkblink/vakya2.0)
39
+ - **Based on**: [IndicF5](https://github.com/AI4Bharat/IndicF5) by AI4Bharat (IIT Madras)
40
+ - **Model Size**: 0.4B parameters
41
+ - **Sample Rate**: 24000 Hz
42
+ - **Training Data**: 1417 hours of high-quality speech
43
+ - **License**: MIT License
44
 
45
+ ## 💡 Tips for Best Results
 
 
 
 
 
 
 
 
 
46
 
47
+ - Keep reference audio clips short (<15 seconds) for best results
48
+ - Use clear, high-quality reference audio
49
+ - Provide reference text when possible for better voice matching
50
+ - The model works best with native speakers of the target language
51
 
52
+ ## ⚠️ Terms of Use
53
 
54
+ - You must have explicit permission to clone voices
55
+ - Unauthorized voice cloning is strictly prohibited
56
+ - Any misuse of this model is the responsibility of the user
57
+ - This model is for research and educational purposes
58
 
59
+ ## 🔗 Links
60
 
61
+ - **Model Repository**: [ashishkblink/vakya2.0](https://huggingface.co/ashishkblink/vakya2.0)
62
+ - **GitHub**: [ashishkblink/vakya](https://github.com/ashishkblink/vakya)
63
+ - **IndicF5**: [AI4Bharat/IndicF5](https://github.com/AI4Bharat/IndicF5)
64
 
65
+ ## 🙏 Acknowledgments
66
+
67
+ This model is based on **IndicF5** developed by AI4Bharat (IIT Madras).
68
 
69
  ---
70
 
71
+ **Vakya** - Bringing voices to Indian languages 🎙️
72
+