TokenVerse: Towards Unifying Speech and NLP Tasks via Transducer-based ASR
Shashi Kumar, Srikanth Madikeri, Juan Pablo Zuluaga Gomez, Iuliia Thorbecke, Esaú Villatoro-tello, Sergio Burdisso, Petr Motlicek, Karthik Pandia D S
Abstract
In traditional conversational intelligence from speech, a cascaded pipeline is used, involving tasks such as voice activity detection, diarization, transcription, and subsequent processing with different NLP models for tasks like semantic endpointing and named entity recognition (NER). Our paper introduces TokenVerse, a single Transducer-based model designed to handle multiple tasks. This is achieved by integrating task-specific tokens into the reference text during ASR model training, streamlining the inference and eliminating the need for separate NLP models. In addition to ASR, we conduct experiments on 3 different tasks: speaker change detection, endpointing, and NER. Our experiments on a public and a private dataset show that the proposed method improves ASR by up to 7.7% in relative WER while outperforming the cascaded pipeline approach in individual task performance. Our code is publicly available: https://github.com/idiap/tokenverse-unifying-speech-nlp
BibTeX
@inproceedings{kumar-etal-2024-tokenverse,
title = "{T}oken{V}erse: Towards Unifying Speech and {NLP} Tasks via Transducer-based {ASR}",
author = "Kumar, Shashi and
Madikeri, Srikanth and
Zuluaga Gomez, Juan Pablo and
Thorbecke, Iuliia and
Villatoro-tello, Esa{\'u} and
Burdisso, Sergio and
Motlicek, Petr and
S, Karthik Pandia D and
Ganapathiraju, Aravind",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.1167/",
doi = "10.18653/v1/2024.emnlp-main.1167",
pages = "20988--20995"
}