analiza aiqbench
This study demonstrates the viability of fine-tuning the Whisper foundation model for Automatic Speech Recognition in Baniwa, a low-resource indigenous language, establishing a critical baseline for preserving linguistic heritage through AI. By adapting a multilingual large-scale model to just 0.54 hours of data, researchers prove that supervised learning can yield functional ASR systems even with extremely limited speech corpora. This work highlights the potential of transfer learning to bridge the digital divide for endangered languages, though future iterations will require larger datasets and specialized post-processing to reduce current error rates.
Fine-Tuning Whisper for Automatic Speech Recognition in Baniwa: A Preliminary Study
arXiv:2608.26060v1 Announce Type: new Abstract: Automatic Speech Recognition (ASR) technologies have achieved remarkable performance in recent years through the use of large multilingual foundation models. However, most advances remain concentrated on high-resource languages, while indigenous languages continue to suffer from a lack of speech resources and language technologies. This work presents a preliminary study on the adaptation of Whisper for Automatic Speech Recognition in Baniwa, an indigenous Arawakan language spoken in Brazil, Colombia, and Venezuela. The experiments were conducted using a corpus of 1,373 manually transcribed recordings obtained from a linguistic documentation project. The corpus contains approximately 0.54 hours of speech and consists primarily of isolated words and short elicited utterances. The Whisper Small model was fine-tuned using supervised learning and evaluated using Word Error Rate (WER) and Character Error Rate (CER). The best model achieved a WER of 37.5% and a CER of 7.45%, demonstrating that multilingual foundation models can be successfully adapted to extremely low-resource indigenous languages. The results establish an initial baseline for Baniwa Automatic Speech Recognition and provide a foundation for future research involving larger datasets, language-specific adaptation strategies, and post-processing techniques.
Abstract:Automatic Speech Recognition (ASR) technologies have achieved remarkable performance in recent years through the use of large multilingual foundation models. However, most advances remain concentrated on high-resource languages, while indigenous languages continue to suffer from a lack of speech resources and language technologies. This work presents a preliminary study on the adaptation of Whisper for Automatic Speech Recognition in Baniwa, an indigenous Arawakan language spoken in Brazil, Colombia, and Venezuela. The experiments were conducted using a corpus of 1,373 manually transcribed recordings obtained from a linguistic documentation project. The corpus contains approximately 0.54 hours of speech and consists primarily of isolated words and short elicited utterances. The Whisper Small model was fine-tuned using supervised learning and evaluated using Word Error Rate (WER) and Character Error Rate (CER). The best model achieved a WER of 37.5% and a CER of 7.45%, demonstrating that multilingual foundation models can be successfully adapted to extremely low-resource indigenous languages. The results establish an initial baseline for Baniwa Automatic Speech Recognition and provide a foundation for future research involving larger datasets, language-specific adaptation strategies, and post-processing techniques.
12 pages, 3 tables. Preliminary study
Computation and Language (cs.CL); Machine Learning (stat.ML)
arXiv:2608.26060 [cs.CL]
(or arXiv:2608.26060v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.26060 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 285 cuvinte
Subiecte conexe
A Speech Corpus for Mizo Automatic Speech Recognition: Whisper and SraVaani 1.0 Fine-Tuning with Morphology-Aware Evaluation
arXiv · 21 aug. 2026
Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models
arXiv · 21 iul. 2026
A Study of ASR Adaptation and Representation Dimensionality Reduction in Persian Speech Emotion Recognition Using Whisper
arXiv · 7 aug. 2026
Reduce ASR inference costs by 75% with NVIDIA MPS on Amazon EC2
AWS · 27 aug. 2026