How much audio is needed to fully observe a multilingual ASR model’s learned sub-token inventory across languages, and does data disparity in multilingual pre-training affect how these tokens are utilized during inference? We address this question by analyzing Whisper’s decoding behavior during inference across 49 languages. By logging decoding candidate sub-tokens and tracking their cumulative discovery over time, we study the utilization pattern of the model’s sub-token space. Results show that the total number of discovered tokens remains largely independent of a language’s pre-training hours, indicating that data disparity does not strongly influence lexical diversity in the model’s hypothesis space. Sub-token discovery rates follow a consistent exponential saturation pattern across languages, suggesting a stable time window after which additional audio yields minimal new token activation. We refer to this convergence threshold as acoustic saturation time AST. Further analyses of rank–frequency distributions reveal Zipf-like patterns better modeled by a Zipf–Mandelbrot law, and mean sub-token length shows a positive correlation with resource level. Additionally, those metrics show more favorable patterns for languages in the Latin script than those in scripts such as Cyrillic, CJK and Semitic. Together, our study suggests that sub-token utilization during multilingual ASR inference is constrained more by the statistical, typological, and orthographical structure of the speech than by training data scale, providing an empirical basis for more equitable corpus construction and cross-lingual evaluation.

This paper presents a systematic evaluation of racial bias in four major commercial automatic speech recognition (ASR) systems using the Pacific Northwest English (PNWE) corpus. We analyze transcription accuracy across speakers from four ethnic backgrounds (African American, Caucasian American, ChicanX, and Yakama) and examine how sociophonetic variation contributes to differential system performance. We introduce a heuristically-determined Phonetic Error Rate (PER) metric that links recognition errors to specific linguistically motivated variables derived from sociophonetic annotation. Our analysis of eleven sociophonetic features reveals that vowel quality variation, particularly resistance to the low-back merger and pre-nasal merger patterns, is systematically associated with differential error rates across ethnic groups, with the most pronounced effects for African American speakers across all evaluated systems. These findings demonstrate that acoustic modeling of dialectal phonetic variation, rather than lexical or syntactic factors, remains a primary source of bias in commercial ASR systems. The study establishes the PNWE corpus as a valuable resource for bias evaluation in speech technologies and provides actionable guidance for improving ASR performance through targeted representation of sociophonetic diversity in training data.

Interlinear glossed text (IGT) creation remains a major bottleneck in linguistic documentation and fieldwork, particularly for low-resource morphologically rich languages. We present a hybrid automatic glossing pipeline that combines neural sequence labeling with large language model (LLM) post-correction, evaluated on Jungar Tuvan, a low-resource Turkic language. Through systematic ablation studies, we show that retrieval-augmented prompting provides substantial gains over random example selection. We further find that morpheme dictionaries paradoxically hurt performance compared to providing no dictionary at all in most cases, and that performance scales approximately logarithmically with the number of few-shot examples. Most significantly, our two-stage pipeline combining a BiLSTM-CRF model with LLM post-correction yields substantial gains for most models, achieving meaningful reductions in annotation workload. Drawing on these findings, we establish concrete design principles for integrating structured prediction models with LLM reasoning in morphologically complex fieldwork contexts. These principles demonstrate that hybrid architectures offer a promising direction for computationally light solutions to automatic linguistic annotation in endangered language documentation.

In-context learning (ICL) enables ASR models to transcribe unseen languages by conditioning on a handful of audio-transcript pairs at inference time, with no fine-tuning. This is appealing for language documentation, where transcribed data is scarce and recording conditions vary across sessions. We evaluate ICL on Panãra (Northern Jê, Brazil), a language with a complex practical orthography in which diacritics encode phonemic contrasts, across seven fieldwork recordings varying in speaker, narrative, and recording context. We find substantial within-language variation in transcription accuracy unexplained by any single recording-level factor, and show that diacritics are a systematic bottleneck with pronounced differences across diacritic types. An orthographic manipulation experiment further shows that how diacritics are represented in context transcriptions substantially affects model performance. These results highlight orthographic complexity and recording-level variation as key practical challenges for ICL-assisted fieldwork transcription.