mlsanityData to 14 Oct 2026

GigaAM Multilingual: Foundation Model for Underrepresented Languages

arXiv 2607.10371 ↗First posted 11 Jul 2026eess.AS cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Despite recent scaling successes, multilingual ASR performance remains highly uneven, with long-tail languages suffering from severe data scarcity. This work addresses the challenge of building robust foundation models for underrepresented Central Asian languages (Kazakh, Kyrgyz, Uzbek). We present GigaAM Multilingual, a Conformer encoder pre-trained on 2M hours of audio using a HuBERT-style objective. Crucially, we introduce a cluster-level data balancing strategy during pre-training and a domain-aware sampling method during fine-tuning to mitigate head-language dominance. In controlled comparisons, our approach outperforms strong open pretrained encoders (Whisper Large v3, Omnilingual-1B) on target languages, achieving significant gains on spontaneous speech while maintaining efficiency. We release the foundation encoder and ASR model, offering a proven recipe for effective multilingual adaptation under realistic data imbalance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 26 of its 35 references that are in the map. First posted on arXiv 11 Jul 2026.

Builds on

26 of 35 references.

References · 26

  1. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  2. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
  3. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  4. 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,342
  5. 2020mT5: A massively multilingual pre-trained text-to-text transformercited by 1,818
  6. 2019Common Voice: A Massively-Multilingual Speech Corpuscited by 1,384
  7. 2021XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scalecited by 601
  8. 2022FLEURS: Few-shot Learning Evaluation of Universal Representations of Speechcited by 469
  9. 2023Scaling Speech Technology to 1,000+ Languagescited by 411
  10. 2023Google USM: Scaling Automatic Speech Recognition Beyond 100 Languagescited by 231
  11. 2018TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptationcited by 226
  12. 2020SLURP: A Spoken Language Understanding Resource Packagecited by 198
14 more
  1. 2021The People's Speech: A Large-Scale Diverse English Speech Recognition Dataset for Commercial Usagecited by 101
  2. 2021SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognitioncited by 78
  3. 2024mHuBERT-147: A Compact Multilingual HuBERT Modelcited by 66
  4. 2024YODAS: Youtube-Oriented Dataset for Audio and Speechcited by 60
  5. 2024GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinementcited by 34
  6. 2025Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languagescited by 34
  7. 2025OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaningcited by 27
  8. 2021KazakhTTS: An Open-Source Kazakh Text-to-Speech Synthesis Datasetcited by 13
  9. 2024On the Effects of Heterogeneous Data Sources on Speech-to-Text Foundation Modelscited by 10
  10. 2021Golos: Russian Dataset for Speech Researchcited by 9
  11. 2021A Toolbox for Construction and Analysis of Speech Datasetscited by 5
  12. 2021USC: An Open-Source Uzbek Speech Corpus and Initial Speech Recognition Experimentscited by 4
  13. 2025The ML-SUPERB 2.0 Challenge: Towards Inclusive ASR Benchmarking for All Language Varietiescited by 4
  14. 2025GigaAM: Efficient Self-Supervised Learner for Speech Recognitioncited by 3

Built on by

Not cited yet.