# Towards Fluent Translations from Disfluent Speech

arXiv: https://arxiv.org/abs/1811.03189 · HTML: https://mlsanity.com/paper/1811.03189/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Nov 2018.

## Built on by (most cited first, 7 of 7)

- [STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation](https://mlsanity.com/paper/2203.10426/) (2022)
- [Learning Shared Semantic Space for Speech-to-Text Translation](https://mlsanity.com/paper/2105.03095/) (2021)
- [Streaming Joint Speech Recognition and Disfluency Detection](https://mlsanity.com/paper/2211.08726/) (2022)
- [DISCO: A Large Scale Human Annotated Corpus for Disfluency Correction in Indo-European Languages](https://mlsanity.com/paper/2310.16749/) (2023)
- [Improving Distinction between ASR Errors and Speech Disfluencies with Feature Space Interpolation](https://mlsanity.com/paper/2108.01812/) (2021)
- [Aligning Stuttered-Speech Research with End-User Needs: Scoping Review, Survey, and Guidelines](https://mlsanity.com/paper/2604.20535/) (2026)
- [The Role of Disfluencies in Speech Translation](https://mlsanity.com/paper/2608.02138/) (2026)

## Abstract (arXiv)

When translating from speech, special consideration for conversational speech phenomena such as disfluencies is necessary. Most machine translation training data consists of well-formed written texts, causing issues when translating spontaneous speech. Previous work has introduced an intermediate step between speech recognition (ASR) and machine translation (MT) to remove disfluencies, making the data better-matched to typical translation text and significantly improving performance. However, with the rise of end-to-end speech translation systems, this intermediate step must be incorporated into the sequence-to-sequence architecture. Further, though translated speech datasets exist, they are typically news or rehearsed speech without many disfluencies (e.g. TED), or the disfluencies are translated into the references (e.g. Fisher). To generate clean translations from disfluent speech, cleaned references are necessary for evaluation. We introduce a corpus of cleaned target data for the Fisher Spanish-English dataset for this task. We compare how different architectures handle disfluencies and provide a baseline for removing disfluencies in end-to-end translation.