---
license: apache-2.0
pipeline_tag: audio-to-audio
---

LavaSR is a novel 50MB BWE(bandwidth extension) model along with the UL-UNAS denoiser.

### Details
* **Model Size:** 50mb for pytorch version.
* **Input Rate:** Any from 8-48khz.
* **Output Rate:** 48kHz
* **Inference Speed:** 10-50x realtime on CPU and 400-4000x realtime depending on GPU.

### Use cases
- Restore low quality audio datasets
- Enhance TTS or ASR model quality.
- Upscale poor quality voice calls.

### Benchmark Comparison

| Model | Speed on GPU(bs=1) | Size | Input range| Quality |
| :--- | :--- | :--- | :--- | :--- |
| **LavaSR** | **4000x** | **50MB** | **Any from 8-48khz** | **High** |
| AudioSR | < 1x realtime | ~3gb+ | ~2-16khz | High |
| AP-BWE(previous formal fastest) | < 400x realtime | ~200MB+ | 8khz/12khz/16khz | Medium |
| NovaSR(previous informal fastest) | <3600x realtime | ~50KB+ | 16khz | Low |

### Usage
Usage instructions can be found here: https://github.com/ysharma3501/LavaSR

### Final notes

The model and code are licensed under the Apache-2.0 license. See LICENSE for details.

Stars/Likes would be appreciated, thank you.

Email: yatharthsharma3501@gmail.com