MOVER: Combining Multiple Meeting Recognition Systems

TL;DR

MOVER is a novel system that effectively combines diverse meeting recognition outputs, including diarization and ASR, to improve accuracy in complex multi-system scenarios.

Contribution

It introduces the first method capable of merging outputs from meeting recognition systems with different diarization and ASR components.

Findings

Achieves 9.55% relative tcpWER reduction on CHiME-8 DASR task.

Achieves 8.51% relative tcpWER reduction on NOTSOFAR-1 multi-channel task.

Demonstrates successful integration of diverse system outputs for improved recognition accuracy.

Abstract

In this paper, we propose Meeting recognizer Output Voting Error Reduction (MOVER), a novel system combination method for meeting recognition tasks. Although there are methods to combine the output of diarization (e.g., DOVER) or automatic speech recognition (ASR) systems (e.g., ROVER), MOVER is the first approach that can combine the outputs of meeting recognition systems that differ in terms of both diarization and ASR. MOVER combines hypotheses with different time intervals and speaker labels through a five-stage process that includes speaker alignment, segment grouping, word and timing combination, etc. Experimental results on the CHiME-8 DASR task and the multi-channel track of the NOTSOFAR-1 task demonstrate that MOVER can successfully combine multiple meeting recognition systems with diverse diarization and recognition outputs, achieving relative tcpWER improvements of 9.55 % and 8.51 % over the state-of-the-art systems for both tasks.