Downloads · 30 days
0
TinySR
MihaiPopa-1/TinySR
TinySR is a audio-to-audio model from MihaiPopa-1. Use it for the audio-to-audio task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
TinySR was the world's smallest super-resolution model (for audio) that I trained!
Access
Public
Updated May 28, 2026
Repo size
164 KB
Likes
4
Public
Repo makeup
Click a slice to open those files.
.js69.4 KB · 28%
At a glance
- Task
- Audio-to-Audio
- License
- apache-2.0
- Access
- Public
- Created
- Dec 29, 2025
- Updated
- May 28, 2026
- SHA
- 11588363
Model card
From the Hugging Face model README
TinySR Readme!
TinySR was the world's smallest super-resolution model (for audio) that I trained!
It's not the world's smallest anymore! The new world's smallest is TinySR 2 and I managed to break the world record!
It's only 57.6 KB in size and it can easily run on a CPU or GPU.
Details
- Input Sample Rate: 16 kHz
- Output Sample Rate: 48 kHz
- Size: Only 57.6 KB
The model works on mono (and stereo, by upscaling both channels) audio!
Training
I used just 46 minutes of instrumental music generated with Mubert, without using any other type of audio!
Benchmark Comparison
| Model | By Myself? | Size |
|---|---|---|
| TinySR (MihaiPopa-1/TinySR) | ✅ | 57.6KB |
| FlashSR (YatharthS/FlashSR) | ❌ | 2MB (PyTorch) / 500KB (ONNX) |
| AudioSR (haoheliu/audiosr_basic) | ❌ | 6.18GB |
Other things
I used a 16-second clip generated with Suno (with vocals) and the model worked!
It's not exactly like the ground truth (and not perfect), but it's better than I thought!
- Task
- Audio-to-Audio
- License
- apache-2.0
- Created
- Dec 29, 2025
- Updated
- May 28, 2026