# Building High-accuracy Multilingual ASR with Gated Language Experts and Curriculum Training

arXiv: https://arxiv.org/abs/2303.00786 · HTML: https://mlsanity.com/paper/2303.00786/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 33 references that are in the map.
- First posted on arXiv 1 Mar 2023.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Listen, Attend and Spell](https://mlsanity.com/paper/1508.01211/) (2015)
- [Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss](https://mlsanity.com/paper/2002.02562/) (2020)
- [Streaming End-to-end Speech Recognition For Mobile Devices](https://mlsanity.com/paper/1811.06621/) (2018)
- [Recent Advances in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/2111.01690/) (2021)
- [Massively Multilingual ASR: 50 Languages, 1 Model, 1 Billion Parameters](https://mlsanity.com/paper/2007.03001/) (2020)
- [Developing Real-time Streaming Transformer Transducer for Speech Recognition on Large-scale Dataset](https://mlsanity.com/paper/2010.11395/) (2020)
- [Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Model](https://mlsanity.com/paper/1909.05330/) (2019)
- [Multilingual Speech Recognition With A Single End-To-End Model](https://mlsanity.com/paper/1711.01694/) (2017)
- [Transformer-Transducer: End-to-End Speech Recognition with Self-Attention](https://mlsanity.com/paper/1910.12977/) (2019)
- [Improving RNN Transducer Modeling for End-to-End Speech Recognition](https://mlsanity.com/paper/1909.12415/) (2019)
- [Bytes are All You Need: End-to-End Multilingual Speech Recognition and Synthesis with Bytes](https://mlsanity.com/paper/1811.09021/) (2018)
- [Massively Multilingual Adversarial Speech Recognition](https://mlsanity.com/paper/1904.02210/) (2019)
- [A Configurable Multilingual Model is All You Need to Recognize All Languages](https://mlsanity.com/paper/2107.05876/) (2021)
- [Transformer-Transducers for Code-Switched Speech Recognition](https://mlsanity.com/paper/2011.15023/) (2020)
- [Streaming End-to-End Multilingual Speech Recognition with Joint Language Identification](https://mlsanity.com/paper/2209.06058/) (2022)
- [From English to More Languages: Parameter-Efficient Model Reprogramming for Cross-Lingual Speech Recognition](https://mlsanity.com/paper/2301.07851/) (2023)
- [Streaming End-to-End Bilingual ASR Systems with Joint Language Identification](https://mlsanity.com/paper/2007.03900/) (2020)
- [Mixtures of Deep Neural Experts for Automated Speech Scoring](https://mlsanity.com/paper/2106.12475/) (2021)
- [UML: A Universal Monolingual Output Layer for Multilingual ASR](https://mlsanity.com/paper/2302.11186/) (2023)

## Built on by (most cited first, 4 of 4)

- [LoRA-Whisper: Parameter-Efficient and Extensible Multilingual ASR](https://mlsanity.com/paper/2406.06619/) (2024)
- [Efficient Adapter Finetuning for Tail Languages in Streaming Multilingual ASR](https://mlsanity.com/paper/2401.08992/) (2024)
- [Language-Aware Prompt Tuning for Parameter-Efficient Seamless Language Expansion in Multilingual ASR](https://mlsanity.com/paper/2506.21577/) (2025)
- [Configurable Multilingual ASR with Speech Summary Representations](https://mlsanity.com/paper/2410.04478/) (2024)

## Abstract (arXiv)

We propose gated language experts and curriculum training to enhance multilingual transformer transducer models without requiring language identification (LID) input from users during inference. Our method incorporates a gating mechanism and LID loss, enabling transformer experts to learn language-specific information. By combining gated transformer experts with shared transformer layers, we construct multilingual transformer blocks and utilize linear experts to effectively regularize the joint network. The curriculum training scheme leverages LID to guide the gated experts in improving their respective language performance. Experimental results on a bilingual task involving English and Spanish demonstrate significant improvements, with average relative word error reductions of 12.5% and 7.3% compared to the baseline bilingual and monolingual models, respectively. Notably, our method achieves performance comparable to the upper-bound model trained and inferred with oracle LID. Extending our approach to trilingual, quadrilingual, and pentalingual models reveals similar advantages to those observed in the bilingual models, highlighting its ease of extension to multiple languages.