Unified Modeling of Multi-Talker Overlapped Speech Recognition and Diarization with a Sidecar Separator
This addresses the challenge of overlapped speech for speech processing systems, but it is incremental as it builds on an existing Sidecar method.
The paper tackles multi-talker overlapped speech recognition and diarization by proposing a unified modeling method that extends a Sidecar separator to include a diarization branch, achieving better ASR results on LibriMix and LibriSpeechMix datasets and acceptable diarization results on CALLHOME with minimal parameter overhead.
Multi-talker overlapped speech poses a significant challenge for speech recognition and diarization. Recent research indicated that these two tasks are inter-dependent and complementary, motivating us to explore a unified modeling method to address them in the context of overlapped speech. A recent study proposed a cost-effective method to convert a single-talker automatic speech recognition (ASR) system into a multi-talker one, by inserting a Sidecar separator into the frozen well-trained ASR model. Extending on this, we incorporate a diarization branch into the Sidecar, allowing for unified modeling of both ASR and diarization with a negligible overhead of only 768 parameters. The proposed method yields better ASR results compared to the baseline on LibriMix and LibriSpeechMix datasets. Moreover, without sophisticated customization on the diarization task, our method achieves acceptable diarization results on the two-speaker subset of CALLHOME with only a few adaptation steps.