NLP

Dual-decoder Transformer for Joint Automatic Speech Recognition and Multilingual Speech Translation

November 9, 2020

Abstract

We introduce dual-decoder Transformer, a new model architecture that jointly performs automatic speech recognition (ASR) and multilingual speech translation (ST). Our models are based on the original Transformer architecture (Vaswani et al., 2017) but consist of two decoders, each responsible for one task (ASR or ST). Our major contribution lies in how these decoders interact with each other: one decoder can attend to different information sources from the other via a dual-attention mechanism. We propose two variants of these architectures corresponding to two different levels of dependencies between the decoders, called the parallel and cross dual-decoder Transformers, respectively. Extensive experiments on the MuST-C dataset show that our models outperform the previously-reported highest translation performance in the multilingual settings, and outperform as well bilingual one-to-one results. Furthermore, our parallel models demonstrate no trade-off between ASR and ST compared to the vanilla multi-task architecture. Our code and pre-trained models are available at https://github.com/formiel/speech-translation.

Download the Paper

AUTHORS

Written by

Hang Le

Juan Pino

Changhan Wang

Jiatao Gu

Didier Schwab

Laurent Besacier

Publisher

International Conference on Computational Linguistics (COLING) 2020

Research Topics

Natural Language Processing

Related Publications

March 17, 2026

NLP

Omnilingual MT: Machine Translation for 1,600 Languages

Omnilingual MT Team, Niyati Bafna, Ioannis Tsiamas, Mark Duppenthaler, Albert Ventayol-Boada, Alexandre Mourachko, Andrea Caciolai, Arina Turkatenko, Artyom Kozhevnikov, Belen Alastruey, Charles-Eric Saint-James, Chierh CHENG, Christophe Ropers, Cynthia Gao, David Dale, Edan Toledo, Eduardo Sánchez, Gabriel Mejia Gonzalez, Holger Schwenk, Jean Maillard, Joe Chuang, João Maria Janeiro, Kevin Heffernan, Marta R. Costa-jussa, Mary Williamson, Nate Ekberg, Paul-Ambroise Duquenne, Pere Lluís Huguet Cabot, Rashel Moritz, Shireen Yates, Surya Parimi

March 17, 2026

Read the Paper

March 17, 2026

Speech & Audio

Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech

Omnilingual SONAR Team, Ioannis Tsiamas, Yen Meng, Vivek Iyer, Guillem Ramirez, Jaehyeong Jo, Alexandre Mourachko, Yu-An Chung, Artyom Kozhevnikov, Belen Alastruey, Christophe Ropers, David Dale, Holger Schwenk, João Maria Janeiro, Kevin Heffernan, Loic Barrault, Marta R. Costa-jussa, Paul-Ambroise Duquenne, Pere Lluís Huguet Cabot

March 17, 2026

Read the Paper

February 27, 2026

Human & Machine Intelligence

Unified Vision–Language Modeling via Concept Space Alignment

Yifu Qiu, Holger Schwenk, Paul-Ambroise Duquenne

February 27, 2026

Read the Paper

November 10, 2025

Speech & Audio

Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages

Omnilingual ASR team, Skyler Wang, Ife Adebara, Michael Auli, Kaushik Ram Sadagopan, Zheng-Xin Yong, Albert Ventayol-Boada, Alexandre Mourachko, Alexander Erben, Yu-An Chung, Arina Turkatenko, Artyom Kozhevnikov, Caley Drooff, Can Balioglu, Chierh Cheng, Christophe Ropers, Cynthia Gao, Gabriel Mejia Gonzalez, Gil Keren, Jean Maillard, Joe Chuang, Kehan Lyu, Kevin Chan, Mark Duppenthaler, Mary Williamson, Matthew Setzler, Paul-Ambroise Duquenne, Rashel Moritz, Safiyyah Saleem, Sagar Miglani, Shireen Yates, Vineel Pratap, Yen Meng

November 10, 2025

Read the Paper

October 31, 2019