Mitigating Speaker Leakage in Cascaded Multi-talker ASR with Diarization-based Transcript Correction

Hermann Yepdjio Nkouanga, Minwei Luo, Maggie Wigness +1

While cascaded multi-talker ASR (MT-ASR) leverages state-of-the-art foundation models, its performance is often capped by speaker leakage during separation. Prior correction strate…