# What shall we do with an hour of data? Speech recognition for the un- and under-served languages of Common Voice

arXiv: https://arxiv.org/abs/2105.04674 · HTML: https://mlsanity.com/paper/2105.04674/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 10 references that are in the map.
- First posted on arXiv 10 May 2021.

## Builds on (most cited first, 3 of 3)

- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [Few-Shot Keyword Spotting in Any Language](https://mlsanity.com/paper/2104.01454/) (2021)

## Built on by (most cited first, 5 of 5)

- [Phonetic Segmentation of the UCLA Phonetics Lab Archive](https://mlsanity.com/paper/2403.19509/) (2024)
- [The Makerere Radio Speech Corpus: A Luganda Radio Corpus for Automatic Speech Recognition](https://mlsanity.com/paper/2206.09790/) (2022)
- [Whisper-LM: Improving ASR Models with Language Models for Low-Resource Languages](https://mlsanity.com/paper/2503.23542/) (2025)
- [Building and curating conversational corpora for diversity-aware language science and technology](https://mlsanity.com/paper/2203.03399/) (2022)
- [A baseline model for computationally inexpensive speech recognition for Kazakh using the Coqui STT framework](https://mlsanity.com/paper/2107.10637/) (2021)

## Abstract (arXiv)

This technical report describes the methods and results of a three-week sprint to produce deployable speech recognition models for 31 under-served languages of the Common Voice project. We outline the preprocessing steps, hyperparameter selection, and resulting accuracy on official testing sets. In addition to this we evaluate the models on multiple tasks: closed-vocabulary speech recognition, pre-transcription, forced alignment, and key-word spotting. The following experiments use Coqui STT, a toolkit for training and deployment of neural Speech-to-Text models.