Mega-ASR
Mega-ASR is a robust automatic speech recognition system designed for real-world audio with severe acoustic degradation.
audio → text · made by zhifeixie
Sold by
Nobody in the catalogue publishes a price for this yet.
About
Mega-ASR — a text model from zhifeixie.
It takes audio and returns text. It was published in May 2026. The catalogue files it under transcribe. Its weights are published under APACHE-2.0, so you may run it on your own machine, or buy it from whoever serves it cheapest.
Every current figure
- Maker
- zhifeixie
- Register
- model
- Takes
- audio
- Returns
- text
- Published
- May 2026
- Licence
- apache-2.0
Known as 1 name
zhifeixie/Mega-ASR