2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

Paper Detail

Paper IDSPE-55.2
Paper Title JOINT ASR AND LANGUAGE IDENTIFICATION USING RNN-T: AN EFFICIENT APPROACH TO DYNAMIC LANGUAGE SWITCHING
Authors Surabhi Punjabi, Harish Arsikere, Zeynab Raeesy, Chander Chandak, Nikhil Bhave, Ankish Bansal, Markus Muller, Sergio Murillo, Ariya Rastrow, Andreas Stolcke, Jasha Droppo, Sri Garimella, Roland Maas, Mat Hans, Athanasios Mouchtaris, Siegfried Kunzmann, Amazon, India
SessionSPE-55: Language Identification and Low Resource Speech Recognition
LocationGather.Town
Session Time:Friday, 11 June, 14:00 - 14:45
Presentation Time:Friday, 11 June, 14:00 - 14:45
Presentation Poster
Topic Speech Processing: [SPE-MULT] Multilingual Recognition and Identification
IEEE Xplore Open Preview  Click here to view in IEEE Xplore
Virtual Presentation  Click here to watch in the Virtual Conference
Abstract Conventional dynamic language switching enables seamless multilingual interactions by running several monolingual ASR systems in parallel and triggering the appropriate downstream components using a standalone language identification (LID) service. Since this solution is neither scalable nor cost- and memory-efficient, especially for on-device applications, we propose end-to-end, streaming, joint ASR-LID architectures based on the recurrent neural network transducer framework. Two key formulations are explored: (1) joint training using a unified output space for ASR and LID vocabularies, and (2) joint training viewed as multi-task optimization. We also evaluate the benefit of using auxiliary language information obtained on-the-fly from an acoustic LID classifier. Experiments with the English-Hindi language pair show that: (a) multi-task architectures perform better overall, and (b) the best joint architecture surpasses monolingual ASR (6.4-9.2% word error rate reduction) and acoustic LID (53.9-56.1% error rate reduction) baselines while reducing the overall memory footprint by up to 46%.