Spaces:
Runtime error
Runtime error
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,13 +1,63 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: blue
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
-
license:
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: LFM2-Audio Speech-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: other
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# LFM2-Audio Speech-to-Speech Chat
|
| 14 |
+
|
| 15 |
+
This is a demo of LFM2-Audio-1.5B, Liquid AI's first end-to-end audio foundation model. Built with low-latency in mind, the lightweight LFM2 backbone enables real-time speech-to-speech conversations without sacrificing quality.
|
| 16 |
+
|
| 17 |
+
## Features
|
| 18 |
+
|
| 19 |
+
- **Real-time speech-to-speech**: Talk to the model and get audio responses
|
| 20 |
+
- **Multi-turn conversations**: Maintain context across multiple exchanges
|
| 21 |
+
- **Interleaved text and audio**: See the text transcription while hearing the audio
|
| 22 |
+
|
| 23 |
+
## How to Use
|
| 24 |
+
|
| 25 |
+
1. **Record your voice**: Click the microphone button and speak your message
|
| 26 |
+
2. **Adjust parameters** (optional):
|
| 27 |
+
- Temperature: Controls randomness (higher = more creative)
|
| 28 |
+
- Top-k: Limits sampling to top k tokens
|
| 29 |
+
3. **Generate Response**: Click the button to get the model's response
|
| 30 |
+
4. **Listen & Read**: Hear the audio response and read the text transcription
|
| 31 |
+
|
| 32 |
+
## Parameters
|
| 33 |
+
|
| 34 |
+
- **Temperature**:
|
| 35 |
+
- 0 = Greedy decoding (most deterministic)
|
| 36 |
+
- 1.0 = Default (balanced)
|
| 37 |
+
- 2.0 = Very creative (more random)
|
| 38 |
+
|
| 39 |
+
- **Top-k**:
|
| 40 |
+
- 0 = No filtering
|
| 41 |
+
- 4 = Default (conservative)
|
| 42 |
+
- Higher values = more diversity
|
| 43 |
+
|
| 44 |
+
## Technical Details
|
| 45 |
+
|
| 46 |
+
- Model: LFM2-Audio-1.5B
|
| 47 |
+
- Audio Codec: Mimi (24kHz)
|
| 48 |
+
- Mode: Interleaved generation (optimal for real-time conversations)
|
| 49 |
+
|
| 50 |
+
## Requirements
|
| 51 |
+
|
| 52 |
+
- GPU recommended for real-time performance
|
| 53 |
+
- Microphone access in your browser
|
| 54 |
+
|
| 55 |
+
## Links
|
| 56 |
+
|
| 57 |
+
- [Liquid AI Website](https://www.liquid.ai/)
|
| 58 |
+
- [GitHub Repository](https://github.com/Liquid4All/liquid-audio/)
|
| 59 |
+
- [Model on Hugging Face](https://huggingface.co/LiquidAI/LFM2-Audio-1.5B)
|
| 60 |
+
|
| 61 |
+
## License
|
| 62 |
+
|
| 63 |
+
Licensed under the LFM Open License v1.0
|