PP-MeT: a Real-world Personalized Prompt based Meeting Transcription System

TL;DR

PP-MeT is a real-world personalized prompt based meeting transcription system that leverages pre-trained models for improved speaker attribution and transcription accuracy in challenging multi-speaker environments.

Contribution

The paper introduces PP-MeT, a novel system combining clustering, TS-VAD, and TS-ASR with pre-trained models for enhanced speaker-attributed ASR in real-world scenarios.

Findings

Achieved cp-CER of 11.27% on M2MeT2.0 dataset

Ranked first in fixed and open training conditions

Demonstrated improved generalizability and precision

Abstract

Speaker-attributed automatic speech recognition (SA-ASR) improves the accuracy and applicability of multi-speaker ASR systems in real-world scenarios by assigning speaker labels to transcribed texts. However, SA-ASR poses unique challenges due to factors such as speaker overlap, speaker variability, background noise, and reverberation. In this study, we propose PP-MeT system, a real-world personalized prompt based meeting transcription system, which consists of a clustering system, target-speaker voice activity detection (TS-VAD), and TS-ASR. Specifically, we utilize target-speaker embedding as a prompt in TS-VAD and TS-ASR modules in our proposed system. In constrast with previous system, we fully leverage pre-trained models for system initialization, thereby bestowing our approach with heightened generalizability and precision. Experiments on M2MeT2.0 Challenge dataset show that our system achieves a cp-CER of 11.27% on the test set, ranking first in both fixed and open training conditions.