# Kunnafonidilaw ka Cadeau: an ASR dataset of present-day Bambara

arXiv: https://arxiv.org/abs/2512.19400 · HTML: https://mlsanity.com/paper/2512.19400/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 22 Dec 2025.

## Builds on (most cited first, 10 of 10)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Scaling Speech Technology to 1,000+ Languages](https://mlsanity.com/paper/2305.13516/) (2023)
- [Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition](https://mlsanity.com/paper/2305.05084/) (2023)
- [Speech Recognition with Augmented Synthesized Speech](https://mlsanity.com/paper/1909.11699/) (2019)
- [Efficient Sequence Transduction by Jointly Predicting Tokens and Durations](https://mlsanity.com/paper/2304.06795/) (2023)
- [Using Radio Archives for Low-Resource Speech Recognition: Towards an Intelligent Virtual Assistant for Illiterate Users](https://mlsanity.com/paper/2104.13083/) (2021)
- [Leveraging Synthetic Audio Data for End-to-End Low-Resource Speech Translation](https://mlsanity.com/paper/2406.17363/) (2024)
- [Synthetic Voice Data for Automatic Speech Recognition in African Languages](https://mlsanity.com/paper/2507.17578/) (2025)
- [Cost Analysis of Human-corrected Transcription for Predominately Oral Languages](https://mlsanity.com/paper/2510.12781/) (2025)
- [Dealing with the Hard Facts of Low-Resource African NLP](https://mlsanity.com/paper/2511.18557/) (2025)

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

- [Listen, Attend, Understand: a Regularization Technique for Stable E2E Speech Translation Training on High Variance labels](https://mlsanity.com/paper/2601.01121/) (2026)
- [Building an ASR Solution for Training and Assessing Children's Reading](https://mlsanity.com/paper/2606.31508/) (2026)

## Abstract (arXiv)

We present Kunkado, a 160-hour Bambara ASR dataset compiled from Malian radio archives to capture present-day spontaneous speech across a wide range of topics. It includes code-switching, disfluencies, background noise, and overlapping speakers that practical ASR systems encounter in real-world use. We finetuned Parakeet-based models on a 33.47-hour human-reviewed subset and apply pragmatic transcript normalization to reduce variability in number formatting, tags, and code-switching annotations. Evaluated on two real-world test sets, finetuning with Kunkado reduces WER from 44.47% to 37.12% on one and from 36.07% to 32.33% on the other. In human evaluation, the resulting model also outperforms a comparable system with the same architecture trained on 98 hours of cleaner, less realistic speech. We release the data and models to support robust ASR for predominantly oral languages.