# On the Inductive Bias of Word-Character-Level Multi-Task Learning for Speech Recognition

arXiv: https://arxiv.org/abs/1812.02308 · HTML: https://mlsanity.com/paper/1812.02308/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Nov 2018.

## Built on by (most cited first, 2 of 2)

- [Hierarchical Conditional End-to-End ASR with CTC and Multi-Granular Subword Units](https://mlsanity.com/paper/2110.04109/) (2021)
- [Learning Mutually Informed Representations for Characters and Subwords](https://mlsanity.com/paper/2311.07853/) (2023)

## Abstract (arXiv)

End-to-end automatic speech recognition (ASR) commonly transcribes audio signals into sequences of characters while its performance is evaluated by measuring the word-error rate (WER). This suggests that predicting sequences of words directly may be helpful instead. However, training with word-level supervision can be more difficult due to the sparsity of examples per label class. In this paper we analyze an end-to-end ASR model that combines a word-and-character representation in a multi-task learning (MTL) framework. We show that it improves on the WER and study how the word-level model can benefit from character-level supervision by analyzing the learned inductive preference bias of each model component empirically. We find that by adding character-level supervision, the MTL model interpolates between recognizing more frequent words (preferred by the word-level model) and shorter words (preferred by the character-level model).