# OWLS: Scaling Laws for Multilingual Speech Recognition and Translation Models

arXiv: https://arxiv.org/abs/2502.10373 · HTML: https://mlsanity.com/paper/2502.10373/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Feb 2025.

## Built on by (most cited first, 17 of 17)

- [Scaling Analysis of Interleaved Speech-Text Language Models](https://mlsanity.com/paper/2504.02398/) (2025)
- [POWSM: A Phonetic Open Whisper-Style Speech Foundation Model](https://mlsanity.com/paper/2510.24992/) (2025)
- [Benchmarking Training Paradigms, Dataset Composition, and Model Scaling for Child ASR in ESPnet](https://mlsanity.com/paper/2508.16576/) (2025)
- [Omni-Router: Sharing Routing Decisions in Sparse Mixture-of-Experts for Speech Recognition](https://mlsanity.com/paper/2507.05724/) (2025)
- [OLMoASR: Open Models and Data for Training Robust Speech Recognition Models](https://mlsanity.com/paper/2508.20869/) (2025)
- [SSVD: Structured SVD for Parameter-Efficient Fine-Tuning and Benchmarking under Domain Shift in ASR](https://mlsanity.com/paper/2509.02830/) (2025)
- [The ML-SUPERB 2.0 Challenge: Towards Inclusive ASR Benchmarking for All Language Varieties](https://mlsanity.com/paper/2509.07139/) (2025)
- [TICL: Text-Embedding KNN For Speech In-Context Learning Unlocks Speech Recognition Abilities of Large Multimodal Models](https://mlsanity.com/paper/2509.13395/) (2025)
- [SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition](https://mlsanity.com/paper/2601.12600/) (2026)
- [How Well Do Large-Scale Chemical Language Models Transfer to Downstream Tasks?](https://mlsanity.com/paper/2602.11618/) (2026)
- [How Do Electrocardiogram Models Scale?](https://mlsanity.com/paper/2605.17276/) (2026)
- [Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLMs](https://mlsanity.com/paper/2607.06831/) (2026)
- [Universal Robust Speech Adaptation for Cross-Domain Speech Recognition and Enhancement](https://mlsanity.com/paper/2602.04307/) (2026)
- [Data Scale, Not Latency, Shapes Cross-Lingual Encoder Transfer in Streaming ASR](https://mlsanity.com/paper/2606.24169/) (2026)
- [Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models](https://mlsanity.com/paper/2607.17164/) (2026)
- [MoLGE: Mixture of Language Group Experts for Efficient Scaling of Massively Multilingual Speech Recognition](https://mlsanity.com/paper/2607.24030/) (2026)
- [YODAS v3: Over 1 Million Hours of High-Bandwidth, Stereophonic, Multilingual Speech](https://mlsanity.com/paper/2609.29448/) (2026)

## Abstract (arXiv)

Neural scaling laws offer valuable insights for designing robust sequence processing architectures. While these laws have been extensively characterized in other modalities, their behavior in speech remains comparatively underexplored. In this work, we introduce OWLS, an open-access, reproducible suite of multilingual speech recognition and translation models spanning 0.25B to 18B parameters, with the 18B version being the largest speech model, to the best of our knowledge. OWLS leverages up to 360K hours of public speech data across 150 languages, enabling a systematic investigation into how data, model, and compute scaling each influence performance in multilingual speech tasks. We use OWLS to derive neural scaling laws, showing how final performance can be reliably predicted when scaling. One of our key findings is that scaling enhances performance on low-resource languages/dialects, helping to mitigate bias and improve the accessibility of speech technologies. Finally, we show how OWLS can be used to power new research directions by discovering emergent abilities in large-scale speech models. Model checkpoints will be released on https://huggingface.co/collections/espnet/owls-scaling-laws-for-speech-recognition-and-translation-67ab7f991c194065f057ce8d for future studies.