# Improving End-to-end Speech Recognition with Pronunciation-assisted Sub-word Modeling

arXiv: https://arxiv.org/abs/1811.04284 · HTML: https://mlsanity.com/paper/1811.04284/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Nov 2018.

## Built on by (most cited first, 10 of 10)

- [Word Alignment by Fine-tuning Embeddings on Parallel Corpora](https://mlsanity.com/paper/2101.08231/) (2021)
- [Recent Advances in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/2111.01690/) (2021)
- [End-to-End Speech Recognition: A Survey](https://mlsanity.com/paper/2303.03329/) (2023)
- [Hierarchical Conditional End-to-End ASR with CTC and Multi-Granular Subword Units](https://mlsanity.com/paper/2110.04109/) (2021)
- [Acoustic Data-Driven Subword Modeling for End-to-End Speech Recognition](https://mlsanity.com/paper/2104.09106/) (2021)
- [A cost minimization approach to fix the vocabulary size in a tokenizer for an End-to-End ASR system](https://mlsanity.com/paper/2406.02563/) (2024)
- [Enhancing CTC-based speech recognition with diverse modeling units](https://mlsanity.com/paper/2406.03274/) (2024)
- [Improving Word Recognition in Speech Transcriptions by Decision-level Fusion of Stemming and Two-way Phoneme Pruning](https://mlsanity.com/paper/2107.12428/) (2021)
- [Byte Pair Encoding Is All You Need For Automatic Bengali Speech Recognition](https://mlsanity.com/paper/2401.15532/) (2024)
- [Transducers with Pronunciation-aware Embeddings for Automatic Speech Recognition](https://mlsanity.com/paper/2404.04295/) (2024)

## Abstract (arXiv)

Most end-to-end speech recognition systems model text directly as a sequence of characters or sub-words. Current approaches to sub-word extraction only consider character sequence frequencies, which at times produce inferior sub-word segmentation that might lead to erroneous speech recognition output. We propose pronunciation-assisted sub-word modeling (PASM), a sub-word extraction method that leverages the pronunciation information of a word. Experiments show that the proposed method can greatly improve upon the character-based baseline, and also outperform commonly used byte-pair encoding methods.