# Evaluating Tibetan Speech Recognition Across Dialects and Adaptation Regimes: A Multi-Granularity Analysis

arXiv: https://arxiv.org/abs/2610.27002 · HTML: https://mlsanity.com/paper/2610.27002/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Multilingual and low-resource ASR, one of 11 directions in Speech and audio language models.
- Builds on 4 of its 20 references that are in the map.
- First posted on arXiv 19 Oct 2026.

## Placed in

- Speech and audio language models / [Multilingual and low-resource ASR](https://mlsanity.com/speech/multilingual-and-low-resource-asr/)

## Builds on (most cited first, 4 of 4)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [TiBERT: Tibetan Pre-trained Language Model](https://mlsanity.com/paper/2205.07303/) (2022)

## Abstract (arXiv)

Tibetan speech recognition is usually evaluated with character or syllable error rates because the writing system does not mark word boundaries. These metrics can rank systems differently and respond differently to adaptation failures. We combine character error rate, syllable error rate, and segmented word error rate in one evaluation framework. We use dictionary-based, neural, and language-model-based segmentation and compare their boundaries with expert annotations. We evaluate five ASR architectures on Lhasa Tibetan, then test zero-shot transfer and target-corpus adaptation on five additional corpus subsets. We also vary LoRA rank and training-data size to examine adaptation failure. The results show that a lower character error does not always lead to a lower syllable or segmented-word error. These ranking differences persist across corpora. Metric correlations also weaken in some unsuccessful adaptation settings. Higher LoRA ranks improve syllable and segmented-word scores but remain worse than full fine-tuning in the tested range. A standardized phonological representation provides another view of spelling errors. These findings support evaluation at several linguistic levels and motivate better segmentation and adaptation methods for low-resource ASR.