Mega-ASR
InstallableRobust automatic speech recognition for challenging real-world audio. Handles noise, far-field, echo, reverberation, and more using a foundation model trained on 2.6M samples across 54 acoustic scenarios.
Community
Loading...
Robust automatic speech recognition for challenging real-world audio. Handles noise, far-field, echo, reverberation, and more using a foundation model trained on 2.6M samples across 54 acoustic scenarios.