Upload read me again
Browse files
README.md
CHANGED
|
@@ -10,16 +10,22 @@ tags:
|
|
| 10 |
# QuarkAudio-HCodec: A Unified Discrete Audio Tokenizer for High-Fidelity, Multitask Audio Generation
|
| 11 |
|
| 12 |
<p align="center">
|
| 13 |
-
<a href="https://arxiv.org/
|
| 14 |
<img src="https://img.shields.io/badge/Paper-ArXiv-red.svg" alt="Paper">
|
| 15 |
</a>
|
| 16 |
-
<a href="https://
|
|
|
|
|
|
|
|
|
|
| 17 |
<img src="https://img.shields.io/badge/Model-Hugging%20Face-yellow.svg" alt="Hugging Face">
|
| 18 |
</a>
|
|
|
|
|
|
|
|
|
|
| 19 |
</p>
|
| 20 |
|
| 21 |
<p align="center">
|
| 22 |
-
<a href="https://arxiv.org/
|
| 23 |
</p>
|
| 24 |
|
| 25 |
> 🔊 **H-Codec**: *A Unified, Dual-Stream Neural Audio Codec with Adaptive Frame Rate and 48kHz Support*
|
|
@@ -31,7 +37,7 @@ tags:
|
|
| 31 |
- ⚙️ **Multi-Sampling Rate (H-Codec-2.0)**: Extends the sampling rate from **16kHz to 48kHz** under a fixed frame rate, significantly improving audio fidelity and high-frequency detail preservation.
|
| 32 |
- 🌍 **Unified Foundation**: Designed as a core component for multimodal LLMs, supporting diverse downstream tasks: TTS, VC, Editing, TTA, SE, and more.
|
| 33 |
|
| 34 |
-
📄 **Paper**: [arXiv:2510.26372](https://arxiv.org/
|
| 35 |
|
| 36 |
---
|
| 37 |
|
|
@@ -91,3 +97,4 @@ pip install -r requirements.txt
|
|
| 91 |
#!/bin/bash
|
| 92 |
python audio_tokenizer.py
|
| 93 |
```
|
|
|
|
|
|
| 10 |
# QuarkAudio-HCodec: A Unified Discrete Audio Tokenizer for High-Fidelity, Multitask Audio Generation
|
| 11 |
|
| 12 |
<p align="center">
|
| 13 |
+
<a href="https://arxiv.org/pdf/2512.20151">
|
| 14 |
<img src="https://img.shields.io/badge/Paper-ArXiv-red.svg" alt="Paper">
|
| 15 |
</a>
|
| 16 |
+
<a href="https://github.com/alibaba/unified-audio/tree/main/QuarkAudio-HCodec/">
|
| 17 |
+
<img src="https://img.shields.io/badge/GitHub-Code-green.svg" alt="GitHub">
|
| 18 |
+
</a>
|
| 19 |
+
<a href="https://huggingface.co/QuarkAudio/QuarkAudio-HCodec/">
|
| 20 |
<img src="https://img.shields.io/badge/Model-Hugging%20Face-yellow.svg" alt="Hugging Face">
|
| 21 |
</a>
|
| 22 |
+
<a href="https://www.modelscope.cn/models/QuarkAudio/QuarkAudio-HCodec/">
|
| 23 |
+
<img src="https://img.shields.io/badge/Model-%20%E9%AD%94%E6%90%AD-orange.svg" alt="ModelScope">
|
| 24 |
+
</a>
|
| 25 |
</p>
|
| 26 |
|
| 27 |
<p align="center">
|
| 28 |
+
<a href="https://arxiv.org/pdf/2512.20151"><img src="HCodec.jpg" width="70%" /></a>
|
| 29 |
</p>
|
| 30 |
|
| 31 |
> 🔊 **H-Codec**: *A Unified, Dual-Stream Neural Audio Codec with Adaptive Frame Rate and 48kHz Support*
|
|
|
|
| 37 |
- ⚙️ **Multi-Sampling Rate (H-Codec-2.0)**: Extends the sampling rate from **16kHz to 48kHz** under a fixed frame rate, significantly improving audio fidelity and high-frequency detail preservation.
|
| 38 |
- 🌍 **Unified Foundation**: Designed as a core component for multimodal LLMs, supporting diverse downstream tasks: TTS, VC, Editing, TTA, SE, and more.
|
| 39 |
|
| 40 |
+
📄 **Paper**: [arXiv:2510.26372](https://arxiv.org/pdf/2512.20151) | 🤗 **Model**: [Hugging Face Spaces](https://huggingface.co/QuarkAudio/QuarkAudio-HCodec/)
|
| 41 |
|
| 42 |
---
|
| 43 |
|
|
|
|
| 97 |
#!/bin/bash
|
| 98 |
python audio_tokenizer.py
|
| 99 |
```
|
| 100 |
+
|