|
|
--- |
|
|
title: Audio LLM |
|
|
emoji: 🌍 |
|
|
colorFrom: yellow |
|
|
colorTo: green |
|
|
sdk: gradio |
|
|
sdk_version: 5.49.1 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
Check out the configuration reference at [Spaces Config Reference](https://huggingface.co/docs/hub/spaces-config-reference) |
|
|
|
|
|
--- |
|
|
|
|
|
# 🎧 LFM2-Audio-1.5B — Multimodal Chatbot |
|
|
|
|
|
An interactive Hugging Face Space built using **[LiquidAI/LFM2-Audio-1.5B](https://huggingface.co/LiquidAI/LFM2-Audio-1.5B)** — |
|
|
a next-generation **audio-language foundation model** that understands and generates both **text and audio**. |
|
|
|
|
|
## 🚀 Features |
|
|
- 💬 Text input and 🎙️ audio input (record or upload) |
|
|
- 🧠 AI replies in both text and generated audio |
|
|
- 🔁 Persistent multi-turn chat history |
|
|
- ⚡ Runs on GPU-backed Spaces for fast inference |
|
|
|
|
|
## 🧩 Tech Stack |
|
|
- Gradio (UI) |
|
|
- PyTorch |
|
|
- Torchaudio |
|
|
- Liquid-Audio SDK |
|
|
|
|
|
## 🧠 Model Credits |
|
|
Model: [LiquidAI/LFM2-Audio-1.5B](https://huggingface.co/LiquidAI/LFM2-Audio-1.5B) |
|
|
Authors: LiquidAI Team |
|
|
License: LFM Open License v1.0 |
|
|
|
|
|
--- |
|
|
|
|
|
To deploy this Space yourself: |
|
|
1. Create a new **Gradio Space** on [huggingface.co/spaces](https://huggingface.co/spaces) |
|
|
2. Upload `app.py`, `requirements.txt`, and this `README.md` |
|
|
3. Select **GPU hardware** (A10G recommended) |
|
|
4. Done! 🎉 |
