New models for recognizing Russian speech in the Vosk library
The developers of the Vosk library have released new models for recognizing Russian speech: the server model vosk-model-ru-0.22 and the mobile model Vosk-model-small-ru-0.22. These models utilize new speech data and a new neural network architecture, which has improved recognition accuracy by 10-20%. The code and data are distributed under the Apache 2.0 license. Key changes: New data collected from voice assistants significantly enhance the recognition of spoken commands, pronounced […]
