Shree Harsha Bokkahalli Satish

Shree Harsha Bokkahalli Satish, Christoph Minixhofer, Maria Teleki et al.

Speech Large Language Models (SpeechLLMs) process spoken input directly, retaining cues such as accent and perceived gender that were previously removed in cascaded pipelines. This introduces speaker identity dependent variation in responses. We present a large-scale intersectional evaluation of accent and gender bias in three SpeechLLMs using 2,880 controlled interactions across six English accents and two gender presentations, keeping linguistic content constant through voice cloning. Using pointwise LLM-judge ratings, pairwise comparisons, and Best-Worst Scaling with human validation, we detect consistent disparities. Eastern European-accented speech receives lower helpfulness scores, particularly for female-presenting voices. The bias is implicit: responses remain polite but differ in helpfulness. While LLM judges capture the directional trend of these biases, human evaluators exhibit significantly higher sensitivity, uncovering sharper intersectional disparities.

Shree Harsha Bokkahalli Satish, Harm Lameris, Joakim Gustafson et al.

Audio anti-spoofing systems are typically formulated as binary classifiers distinguishing bona fide from spoofed speech. This assumption fails under layered generative processing, where benign transformations introduce distributional shifts that are misclassified as spoofing. We show that phonation-modifying voice conversion and speech restoration are treated as out-of-distribution despite preserving speaker authenticity. Using a multi-class setup separating bona fide, converted, spoofed, and converted-spoofed speech, we analyse model behaviour through self-supervised learning (SSL) embeddings and acoustic correlates. The benign transformations induce a drift in the SSL space, compressing bona fide and spoofed speech and reducing classifier separability. Reformulating anti-spoofing as a multi-class problem improves robustness to benign shifts while preserving spoof detection, suggesting binary systems model the distribution of raw speech rather than authenticity itself.

Shree Harsha Bokkahalli Satish, Maria Teleki, Christoph Minixhofer et al.

SpeechLLMs process spoken language directly from audio, but accent and vocal identity cues can lead to biased behaviour. Current bias evaluations often miss how such bias manifests in end-to-end speech interactions and how users experience it. We distinguish quality-of-service disparities (e.g., off-topic or low-effort responses) from content-level bias in coherent outputs, and examine intersectional effects of accent and perceived gender. In this work, we explore a two-part evaluation approach: (1) a controlled test cohort spanning six accents and two gender presentations, analysed with judge-free prompt-response metrics, and (2) an interactive study design using voice conversion to let users experience identical content through different vocal identities. Across two studies (Interactive, N=24; Observational, N=19), we find that voice conversion increases trust and acceptability for benign responses and encourages perspective-taking, while automated analysis in search of quality-of-service disparities, reveals {accent x gender} disparities in alignment and verbosity across SpeechLLMs. These results highlight voice conversion for probing and experiencing intersectional voice bias while our evaluation suite provides richer bias evaluations for spoken conversational AI.

Shree Harsha Bokkahalli Satish, Gustav Eje Henter, Éva Székely

Recent work in benchmarking bias and fairness in speech large language models (SpeechLLMs) has relied heavily on multiple-choice question answering (MCQA) formats. The model is tasked to choose between stereotypical, anti-stereotypical, or neutral/irrelevant answers given an input speech prompt and an optional text prompt. Such MCQA benchmarks implicitly assume that model performance is consistent across other MCQA tasks, voices, and other task formats such as more realistic, long-form evaluations. In this paper, we probe that assumption. We fine-tune three SpeechLLMs using LoRA adapters to induce specific MCQA behaviours: preference for stereotypical, anti-stereotypical, or neutral/uncertain answers. We then evaluate whether these behaviours generalise to another, distinct MCQA benchmark, and more critically to long-form, creative generation tasks. Our results show that performance on MCQA bias benchmarks fails to reliably predict performances across other MCQA benchmarks, and more importantly across long-form tasks. We conclude that current MCQA bias benchmarks show limited evidence of cross-task generalisation in the speech domain, and also propose an evaluation suite for measuring behaviour transferability in future models and benchmarks.

Harm Lameris, Shree Harsha Bokkahalli Satish, Joakim Gustafson et al.

Recent advances in speech foundation models (SFMs) have enabled the direct processing of spoken language from raw audio, bypassing intermediate textual representations. This capability allows SFMs to be exposed to, and potentially respond to, rich paralinguistic variations embedded in the input speech signal. One under-explored dimension of paralinguistic variation is voice quality, encompassing phonation types such as creaky and breathy voice. These phonation types are known to influence how listeners infer affective state, stance and social meaning in speech. Existing benchmarks for speech understanding largely rely on multiple-choice question answering (MCQA) formats, which are prone to failure and therefore unreliable in capturing the nuanced ways paralinguistic features influence model behaviour. In this paper, we probe SFMs through open-ended generation tasks and speech emotion recognition, evaluating whether model behaviours are consistent across different phonation inputs. We introduce a new parallel dataset featuring synthesized modifications to voice quality, designed to evaluate SFM responses to creaky and breathy voice. Our work provides the first examination of SFM sensitivity to these particular non-lexical aspects of speech perception.

Shree Harsha Bokkahalli Satish, Harm Lameris, Olivier Perrotin et al.

Speech Continuation (SC) is the task of generating a coherent extension of a spoken prompt while preserving both semantic context and speaker identity. Because SC is constrained to a single audio stream, it offers a more direct setting for probing biases in speech foundation models than dialogue does. In this work we present the first systematic evaluation of bias in SC, investigating how gender and phonation type (breathy, creaky, end-creak) affect continuation behaviour. We evaluate three recent models: SpiritLM (base and expressive), VAE-GSLM, and SpeechGPT across speaker similarity, voice quality preservation, and text-based bias metrics. Results show that while both speaker similarity and coherence remain a challenge, textual evaluations reveal significant model and gender interactions: once coherence is sufficiently high (for VAE-GSLM), gender effects emerge on text-metrics such as agency and sentence polarity. In addition, continuations revert toward modal phonation more strongly for female prompts than for male ones, revealing a systematic voice-quality bias. These findings highlight SC as a controlled probe of socially relevant representational biases in speech foundation models, and suggest that it will become an increasingly informative diagnostic as continuation quality improves.