# Who Needs Words? Lexicon-Free Speech Recognition

arXiv: https://arxiv.org/abs/1904.04479 · HTML: https://mlsanity.com/paper/1904.04479/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Apr 2019.

## Built on by (most cited first, 6 of 6)

- [Scaling Speech Technology to 1,000+ Languages](https://mlsanity.com/paper/2305.13516/) (2023)
- [End-to-End Speech Recognition: A Survey](https://mlsanity.com/paper/2303.03329/) (2023)
- [Kaizen: Continuously improving teacher using Exponential Moving Average for semi-supervised speech recognition](https://mlsanity.com/paper/2106.07759/) (2021)
- [Training a Tokenizer for Free with Private Federated Learning](https://mlsanity.com/paper/2203.09943/) (2022)
- [General-Purpose Speech Representation Learning through a Self-Supervised Multi-Granularity Framework](https://mlsanity.com/paper/2102.01930/) (2021)
- [Emphasizing Unseen Words: New Vocabulary Acquisition for End-to-End Speech Recognition](https://mlsanity.com/paper/2302.09723/) (2023)

## Abstract (arXiv)

Lexicon-free speech recognition naturally deals with the problem of out-of-vocabulary (OOV) words. In this paper, we show that character-based language models (LM) can perform as well as word-based LMs for speech recognition, in word error rates (WER), even without restricting the decoding to a lexicon. We study character-based LMs and show that convolutional LMs can effectively leverage large (character) contexts, which is key for good speech recognition performance downstream. We specifically show that the lexicon-free decoding performance (WER) on utterances with OOV words using character-based LMs is better than lexicon-based decoding, both with character or word-based LMs.