NovaSR: Pushing the Limits of Extreme Efficiency in Audio Super-Resolution
This is the model for NovaSR, a tiny 50kb audio upsampling model that upscales muffled 16khz audio into clear and crisp 48khz audio at speeds from 100-3500x realtime.
Audio Samples
Before Processing (16kHz):
After Processing (48kHz):
Details
Model Size:
52kb for pytorch version
Input Rate:
16kHz
Output Rate:
48kHz
Inference Speed:
300-3500x realtime depending on gpu
Mono
Comparisons
Comparisons were done on A100 gpu. Higher realtime means faster processing speeds.
Comparison on CPU are coming soon.
NovaSR huggingface.co is an AI model on huggingface.co that provides NovaSR's model effect (), which can be used instantly with this drbaph NovaSR model. huggingface.co supports a free trial of the NovaSR model, and also provides paid use of the NovaSR. Support call NovaSR model through api, including Node.js, Python, http.
NovaSR huggingface.co is an online trial and call api platform, which integrates NovaSR's modeling effects, including api services, and provides a free online trial of NovaSR, you can try NovaSR online for free by clicking the link below.
NovaSR is an open source model from GitHub that offers a free installation service, and any user can find NovaSR on GitHub to install. At the same time, huggingface.co provides the effect of NovaSR install, users can directly use NovaSR installed effect in huggingface.co for debugging and trial. It also supports api for free installation.