Paper · arXiv 2601.18415
Pisets: A Robust Speech Recognition System for Lectures and Interviews
Ivan Bondarenko, Daniil Grebenkin, Oleg Sedukhin, Mikhail Klementev, Roman Derunets, Lyudmila Budneva
A three-component speech-to-text system combines Wav2Vec2, AST, and Whisper models with curriculum learning and uncertainty modeling to improve transcription accuracy and reduce hallucinations in Russian speech recognition.
Abstract
This work presents a speech-to-text system "Pisets" for scientists and journalists which is based on a three-component architecture aimed at improving speech recognition accuracy while minimizing errors and hallucinations associated with the Whisper model. The architecture comprises primary recognition using Wav2Vec2, false positive filtering via the Audio Spectrogram Transformer (AST), and final speech recognition through Whisper. The implementation of curriculum learning methods and the utilization of diverse Russian-language speech corpora significantly enhanced the system's effectiveness. Additionally, advanced uncertainty modeling techniques were introduced, contributing to further improvements in transcription quality. The proposed approaches ensure robust transcribing of long audio data across various acoustic conditions compared to WhisperX and the usual Whisper model. The source code of "Pisets" system is publicly available at GitHub: https://github.com/bond005/pisets.