Speaker Recognition

A trustworthy and GDPR-compliant deepfake audio detector must base its decisions on acoustic artifacts, not on what is being said or who is speaking. We present a large-scale study of semantic independence for Resemble AI's detector, DETECT-3B-Omni. Using 10,240 audio samples from diverse US English speakers across 30 states, generated through 8 different AI voice-cloning systems, we test whether detection accuracy depends on spoken content (benign versus malicious), speaker gender, speaker age, or speaker region. Using equivalence testing, our results show that the accuracy difference between any two of these groups is at most 2 percentage points, at 99% confidence. The detector therefore identifies AI-generated audio with equivalent accuracy regardless of what the audio says or who the speaker is.

Long-form TV dramas present a formidable challenge for comprehensive video understanding, where deciphering complex storyline often relies on \textbf{speaker recognition}, the task of accurately attributing each spoken utterance to its respective character. In this paper, we advance this field through two primary contributions. (1) We introduce \textbf{DramaSR-532K}, a large-scale benchmark comprising 532K annotated dialogue lines across more than 900 unique characters, necessitating the integration of auditory, linguistic, and visual cues for speaker recognition. (2) We propose \textbf{DramaSR-LRM}, a robust approach built upon a large reasoning model (LRM). DramaSR-LRM is designed to autonomously aggregate contextual evidence via multimodal tool-use, synthesizing diverse inputs to achieve high-fidelity attribution. Experimental results demonstrate that DramaSR-LRM significantly outperforms existing baselines, particularly on short utterances where acoustic biometrics are inherently unreliable. \textit{All the data and code will be made publicly available at the project page: https://www.github.com/198808xc/DramaSR-LRM.}

Turn-taking naturalness is central to full-duplex spoken dialogue systems, yet its automatic evaluation remains limited. Existing evaluations often rely on human judgments or behavior-specific timing metrics, making it difficult to compare heterogeneous timing failures within a unified framework. We propose TurnNat, a likelihood-based framework for automatic turn-taking naturalness evaluation in two-channel spoken dialogue. A causal turn-taking prediction model trained on natural conversations estimates future two-speaker voice-activity states, and the negative log-likelihood (NLL) of the observed future activity measures timing atypicality. TurnNat pools frame-level NLLs over turn-taking boundary units (TBUs) extracted from utterance onsets and offsets, and aggregates mean and tail TBU scores into a dialogue-level naturalness score. We further construct a controlled perturbation benchmark of paired natural and perturbed dialogue clips, validated by human naturalness judgments. Experiments on this benchmark show that TurnNat successfully identifies unnatural turn-taking perturbations across heterogeneous timing failures.

Cross-lingual speaker verification (SV) systems typically exhibit performance degradation when enrollment and test utterances are spoken in different languages. However, standard evaluation protocols confound language mismatch with inter-speaker variability, as evaluation is generally performed with different speakers across languages. In this work, we introduce a bilingual same-speaker evaluation set for five Iberian languages, enabling analysis of cross-lingual SV under constant speaker identity. We apply this setup to a HuBERT-based SV system previously shown to exhibit strong language dependence, and analyze results using the Cross-Lingual Transfer Matrix (CLTM) to study pairwise cross-lingual transfer. Our results show that speaker-related variability accounts for part of the observed degradation, but language mismatch remains the main driver of cross-lingual performance loss. These findings provide a more precise characterization of language dependence in cross-lingual SV.

Automatically detecting stress in speech provides an unobtrusive way to gain insights relevant to behavioral research or clinical assessment. This study investigates the automatic differentiation between a stressful and non-stressful situation, and the prediction of physiological and affective stress responses. Speech data was collected from 50 participants who either completed the Trier Social Stress Test (TSST) or a non-stressful control condition. With a processing pipeline that included speaker diarization and machine learning models, we achieved stress detection performance significantly above a mean baseline. Moreover, relevant physiological and affective stress responses were partially predictable from acoustic-prosodic features. Feature-importance analyses identified the most informative predictors contributing to model performance. The findings demonstrate that speech can serve as a meaningful and unobtrusive indicator of multiple dimensions of the human stress response.

Text-to-speech (TTS) evaluation is an open challenge. While the primary target was "naturalness," recent fidelity gains shifted focus toward "appropriateness" and whether speech is correct for its context. In this work, we examine how perception changes when the expected downstream use varies. We measure the appropriateness and human-likeness of five SOTA TTS systems across five domains: AI assistant, reader, actor, animated character, and spontaneous speaker. Results show appropriateness varies across domains independently of naturalness. While systems shine at reading, expressive domains remain challenging, and optimizing for one can degrade others. Furthermore, naturalness scores tend to penalize stylized speech while rewarding spontaneity. Finally, our study also highlights blind spots in one-size-fits-all evaluation metrics across more expressive domains. We demonstrate that TTS performance is not "solved" but depends on the target domain, requiring context-aware evaluation.

Some neural audio codecs disentangle speech into latent subspaces encoding content, speaker identity, and acoustics, enabling acoustic teleportation and voice conversion. Existing evaluations rely on cross-reconstruction quality, which cannot reliably detect leakage across partitions. We extend a probing based framework to assess disentanglement by regressing room-acoustic parameters (reverberation time, clarity, and direct-to-reverberant ratio) and classifying speaker identity, using the gap between intended and unintended partitions as the disentanglement measure. Applied to an acoustic teleportation codec, we find speaker identity is largely confined to its partition, while acoustics leak into the speech embeddings due to the training objective. Acoustic embeddings blindly estimate room parameters within 0.02 s of supervised baselines, indicating physically meaningful structure emerges without explicit supervision.

UTMOS has become one of the most commonly used deep neural network-based speech quality assessment (SQA) metrics in speech processing research. In this paper, we attack UTMOS to probe its robustness. Starting from high-quality speech samples, we optimize the input in two directions: a score-preserving attack, which degrades perceived quality while maintaining the predicted score, and a quality-preserving attack, which lowers the predicted score while maintaining perceived quality. We consider three input spaces: raw waveform, mel spectrogram with a HiFi-GAN vocoder, and the latent space of EnCodec, a neural audio codec. Experimental results show that score-preserving attacks are effective against UTMOS. Although perfect quality-preserving attacks are more difficult, optimization in the EnCodec latent space provides the best chance of success. These results reveal failure modes of UTMOS and highlight the importance of robustness analysis for DNN-based SQA metrics.

Voice anonymization aims to protect speaker identity while preserving linguistic content and speech usability. However, most anonymization systems are developed on adult speech, leading to degraded performance when applied to child speech. This paper investigates child-centric anonymization by adapting a self-supervised learning (SSL) based anonymization pipeline to the child speech domain. The system is adapted using child speech from the MyST corpus and evaluated under both single-speaker and two-speaker mixture conditions. Experimental results show that child-domain adaptation improves intelligibility and perceptual quality while maintaining strong privacy protection. Extending the approach to multi-speaker further demonstrates that combining target speaker extraction with child-adapted anonymization provides privacy protection while preserving conversational structure. These findings highlight the importance of child-specific adaptation for practical speech anonymization systems.

Poisoning attacks entail attackers intentionally tampering with training data. In this paper, we consider a dirty-label poisoning attack scenario on a speech commands classification system. The threat model assumes that certain utterances from one of the classes (source class) are poisoned by superimposing a trigger on it, and its label is changed to another class selected by the attacker (target class). We propose a filtering defense against such an attack. First, we use DIstillation with NO labels (DINO) to learn unsupervised representations for all the training examples. Next, we use K-means and LDA to cluster these representations. Finally, we keep the utterances with the most repeated label in their cluster for training and discard the rest. For a 10% poisoned source class, we demonstrate a drop in attack success rate from 99.75% to 0.25%. We test our defense against a variety of threat models, including different target and source classes, as well as trigger variations.

Insurance fraud remains costly and operationally difficult, particularly in call-centre workflows where many customer interactions begin at FNOL. While recent fraud detection methods mainly rely on structured data, text, or images, repeated speaker identity across calls remains underused as an investigative signal. This paper presents DG^VoiC, a voice clustering framework for customer verification and cross-profile speaker linking on anonymised real call-centre audio. The approach combines sensitive information-aligned anonymisation, speech-focused preprocessing, sliding-window speaker embedding extraction, and cosine similarity based clustering to identify repeated speakers under real telephony conditions. The method was evaluated on 121 recordings, with a curated reference subset of 56 samples in 22 human-agreed speaker clusters. used for validation. The best configuration achieved 96% AMI, 95% ARI, 98% completeness, 100% homogeneity, and 99% V-measure. These results show that speaker clustering can provide a strong additional signal for fraud investigation by helping analysts verify speaker consistency and surface repeated voices across customers.

The variations in vocal effort range (e.g. whisper, soft, neutral, loud, shout) alter production and speech acoustics, reducing intelligibility and limiting the robustness of any subsequent speech technology. Classification is challenging since effort lies on a continuum, adjacent categories are easily confused, and labeled data remain scarce. Prior SSL approaches with wav2vec2, HuBERT, and AST improve performance on the AVID corpus but still suffer from boundary errors. In this study, we introduce WavLM for the first time in vocal effort classification and benchmark it against wav2vec2 and HuBERT. To address data scarcity, we conduct a systematic study of augmentation strategies, covering RIR convolution, additive noise, time masking, speed perturbation, band-limiting, MixUp, and CutMix. Augmentation consistently improves WavLM, with gains ranging from +0.6% to +1.8% absolute. We further propose Gaussian-neighbor soft labels, which further reduce near-boundary confusions by modeling the vocal effort continuum. Our best system, WavLM-BASE with gradual unfreezing, augmentation, and Gaussian-neighbor soft labels, achieves 78.2% mean accuracy, establishing a new state-of-the-art on AVID.

As multimodal conversational systems increasingly engage in spoken interaction, their ability to navigate paralinguistic social cues has become a critical bottleneck for natural human-AI communication. However, existing evaluations of machine emotional intelligence assess reasoning exclusively through isolated text or passive acoustic perception, overlooking the complex cross-modal reasoning required for active, multi-turn dialogue. We introduce \textsc{SpeechEQ}, a comprehensive framework designed to evaluate the sociolinguistic reasoning of Speech-Language Models (SLMs). The framework includes a validated dataset of 2,265 dialogues across 15 Emotional Quotient (EQ) subscales grounded in EQ-i 2.0 theory, along with a multi-turn evaluation protocol measured by our proposed Spoken EQ (SEQ) score inspired by human EQ assessments. Experiments show limitations in how both existing Speech Emotion Recognition and end-to-end Speech-Language Models understand and apply paralinguistic cues through speech. While end-to-end architectures outperform cascaded systems, \textsc{SpeechEQ} reveals that current multimodal models remain bottlenecked by a text-reliant modality shortcut,'' an alignment-induced safety trap,'' and ``contextual amnesia,'' highlighting the barriers to truly emotionally aware AI. Our benchmark can be accessed at https://huggingface.co/datasets/SpeechEQ/SpeechEQ and demo page at https://binomial14.github.io/speecheq-demo/

Accent conversion and controllability remain fundamental challenges in cross-lingual text-to-speech (TTS), particularly for low-resource and phonetically diverse Indic languages. While recent large language model (LLM)-based TTS systems exhibit strong cross-lingual generalization, they provide limited explicit control over accent characteristics and intensity. In this paper, we propose CrossAccentTTS, a framework that enables both accent control and conversion while preserving speaker identity. Specifically, we introduce an Accent Intensity Controller (AIC) that injects weighted language embeddings into the accent subspace, allowing smooth interpolation between accents and fine-grained modulation of accent strength at inference time. Experiments on the Indic Multilingual and L2-arctic datasets shows that CrossAccent-TTS achieves precise control of accent intensity, outperforming strong baselines in accent similarity and controllability by maintaining speaker similarity and naturalness.

This work investigates the interpretability of a Wav2Vec 2.0based speech intelligibility assessment model for oral and oropharyngeal cancer patients through canonical correlation analysis. By measuring the correlation between the model embeddings and eGeMAPS low-level descriptors (LLDs) as an interpretable reference, we analyze how acoustic information is encoded across the model layers. The analysis is conducted at two levels: individual LLDs layer-wise, and group-level: prosodic, spectral, and voice quality. Results show that the learned representations are most strongly correlated with spectral and prosodic features, with the first MFCC coefficient yielding the highest correlations across all layers. At the group level, spectral and prosodic groups achieve correlations of 0.77 and 0.71 respectively, while voice quality reaches 0.65. Beyond model interpretability, this work also offers practical guidance on acoustic feature selection for pathological speech assessment.

Speaker recognition has advanced rapidly with large-scale training datasets, yet Vietnamese remains under-resourced, with existing corpora limited in scale and acoustic diversity. Most large-scale datasets rely on facial cues to link speech with speaker identities, restricting data collection to recordings where speakers appear on camera. We propose a face-independent dataset construction pipeline and introduce VieSpeaker, a large-scale Vietnamese speaker recognition dataset. Our approach leverages textual metadata and large language model reasoning to infer speaker identities from transcripts and contextual information. VieSpeaker contains approximately 902 hours of speech from 4,715 speakers. Experiments show that models trained on VieSpeaker achieve improved robustness and generalization compared to existing Vietnamese datasets. This work demonstrates the feasibility of face-independent dataset construction and provides a new direction for building large-scale speech resources.

Reviewing recorded interviews for affective cues such as composure, hesitation and agitation is slow and subjective, and cloud services that could automate it require sensitive audio to leave the device. EmotionAI is a fully local Computational Intelligence (CI) pipeline that couples Speech Emotion Recognition (SER) with generative reasoning. Speaker diarisation, Whisper Automatic Speech Recognition (ASR) and a wav2vec2 emotion classifier produce per-segment affective evidence, which is then passed to an adversarial three-model local Large Language Model (LLM) panel for timestamp-grounded and citation-constrained question answering. Zero-shot evaluation on the RAVDESS four-class English subset (n = 672) exposes cross-corpus fragility rather than classifier superiority: the deployed classifier scores 48.8% accuracy, above random (24.9%) and majority (28.6%) baselines but below an in-domain MFCC + logistic-regression comparator (71.0%). The complete pipeline runs in a mean 157 s on CPU (real-time factor approximately 1.33) with zero external calls. The contribution is not state-of-the-art SER but an auditable, privacy-preserving integration of imperfect affective evidence into grounded conversational analysis, together with an honest empirical account of where cross-corpus transfer and human-centred validation still fall short.

Provenance watermarking is increasingly treated as a safeguard for synthetic speech, whether built directly into speech-generation models such as Chatterbox, provided through dedicated techniques such as AudioSeal, or deployed by commercial platforms such as ElevenLabs. We identify a previously uncharacterized liability: when synthetic speech is watermarked and human speech is not, detectors trained alongside latch onto the watermark as a spurious "watermark => fake" shortcut. This single feature yields three coupled failures: generalization degradation (model performance deteriorates on unseen data), strip-to-evade (a watermarked fake escapes once unwatermarked), and mark-to-frame (watermarking a real voice flags it as fake). In a controlled white-box experiment, a watermark-trained detector shows all three (for example, mark-to-frame lifts Equal Error Rate from 16% to 75%). In a black-box test of a commercial API, we show that adding a watermark to real speech disguises it as fake. However, this shortcut is fixable: retraining with the watermark on both classes decorrelates it and restores clean behavior. We release experiment data as a paired clean-versus-watermarked corpus (WASP).

Humans tend to speak louder and clearer in challenging environments, such as noisy conditions or when addressing hearingimpaired listeners, which is called Lombard effect. To simulate this behavior in speech synthesis systems, we introduce a flow-matching based text-to-speech (TTS) model trained with vocal effort and articulation pseudo-labels. The proposed model achieves continuous and disentangled control of vocal effort and articulation, while also enabling word-level emphasis for clarifying specific segments of an utterance. Experimental results show that these control mechanisms effectively improve clarityrelated acoustic features. Furthermore, speech-in-noise experiments demonstrate that our model successfully simulates the intelligibility gains of human clear speech in noisy conditions.

Explaining and understanding the decision-making process of artificial intelligence (AI) systems, particularly those implemented by neural networks, falls within the field of explainable AI (XAI). Analogous to the human attention mechanism, neural networks are assumed to possess their own attention mechanisms that selectively process information during decision-making. This work proposes to study one XAI topic: analysing and visualising the attention mechanisms of neural networks. Our experiments are performed on speaker recognition neural networks that are trained to identify speaker identity from a given utterance. Previous studies have widely used class activation map (CAM)-based methods to analyse and visualise the attention mechanisms of neural networks. Each of these methods produces an attention map for each network input, highlighting which input regions are selectively processed when the speaker recognition network makes decisions. However, the evaluation of attention maps produced by these methods remains largely underexplored. This work systematically reviews an existing attention map evaluation algorithm, establishing key concepts and identifying its shortcomings. On the basis of this existing evaluation algorithm, a new version is then proposed to address the identified shortcomings, called the Modified Randomised Input Sampling for Explanation - Evaluation algorithm (Modified RISE-eval). Using Modified RISE-eval, we evaluate the attention maps produced by two representative CAM-based methods, GradCAM and LayerCAM, applied to a certain speaker recognition network. The evaluation results demonstrate that GradCAM and LayerCAM each exhibit distinct advantages when applied under different experimental conditions in the speaker recognition task.

In this paper, we present Kiwano, an open-source toolkit designed to advance research and evaluation for speaker verification. Kiwano provides a lightweight yet extensible framework built on PyTorch, offering standardized recipes, pretrained models, and integration of several widely used speaker verification architectures. The toolkit emphasizes reproducibility, by delivering transparent training pipelines, unified evaluation protocols and ready-to-use baselines across multiple corpora. Beyond conventional training and inference, Kiwano includes tools for benchmarking, experiment tracking and rapid prototyping of new architectures. To foster community adoption, the toolkit is distributed under the Apache 2.0 license, accompanied by comprehensive documentation and reproducible experiments. By lowering entry barriers and standardizing evaluation practices, Kiwano contributes a valuable resource for both academic research and applied development in speaker verification. The toolkit is publicly available at: https://github.com/kiwano-toolkit/kiwano/

Physics-informed neural operators (PINOs) have recently gained attention as fast numerical simulators with potential for solving inverse problems. This study proposes the first PINO-based method for speech production analysis. The model learns the governing one-dimensional wave equations directly without requiring pre-computed supervised training data. Using vocal tract shape data as input features, we compare the proposed model's predicted f0, glottal volume velocity and sound pressure at the lip for five static vowels to a conventional Runge Kutta/Finite difference approach. With errors of 0.8% for glottal volume flow and 3.2% for speech waveforms, the proposed model enables efficient GPU-parallelized simulation without iterative calculations. We conclude that PINO is a promising approach for fast analysis of speech.

Understanding speaker attributes is crucial for voice-related applications, yet conventional approaches rely on fixed categorical labels, lacking semantic richness and zero-shot generalizability. We propose a novel framework for open-set speaker attribute prediction leveraging Large Language Model (LLM) embeddings to represent attributes in a continuous semantic space. To bridge the cross-modal gap, we introduce a keyword-appending strategy that structures broad semantic representations into a compact, discriminative manifold. Furthermore, we employ a top-k negative loss to establish robust decision boundaries in crowded semantic regions. Experimental results on LibriTTS-P demonstrate that our method outperforms closed-set benchmarks and generalizes effectively to unseen synonyms. Geometric analysis suggests that our strategies regularize the embedding manifold, balancing semantic cohesion with predictive clarity.

Voice reconstruction using Text-to-Speech (TTS) offers a communication method for people with speech disorders, which aims to retain their speaker identity while improving intelligibility. Previous work generally relies on Mean Opinion Score (MOS) to evaluate naturalness and speaker similarity, but this has limited sensitivity and reliability. We propose an evaluation framework with subjective and objective components. Subjectively, we evaluate perceived intelligibility and speaker identity using Best Worst Scaling (BWS) with situational framing. Objectively, we demonstrate that standard measures fail to predict reconstruction success for highly unintelligible speakers, so we introduce a novel dual-reference distributional measure to assess the trade-off between intelligibility and speaker identity. By evaluating the output of 17 zero-shot TTS systems for 193 speakers, we show that our framework provides a reliable and task-aligned approach for assessing voice reconstruction.

Deep neural network-based automatic speaker verification (ASV) systems achieve impressive performance but their embedding representations remain opaque, lacking a structured and perceptually verifiable explanation of the vocal characteristics they encode. Existing approaches either require annotation of speaker attributes or introduce alternative representations whose interpretability is unvalidated with listeners. We propose Listenable Interpretable Speaker Embeddings (LISE), a label-free framework that decomposes pretrained speaker embeddings into a small set of components. This decomposition yields a structured representation that supports the analysis of what information has been encoded by speaker embeddings. LISE preserves ASV performance with negligible EER degradation on x-vector and ECAPA-TDNN. Crucially, the interpretability of these components for human listeners is demonstrated through listening experiments, where participants distinguished speakers with 83.9% accuracy.

As expressive text-to-speech (TTS) and voice conversion (VC) systems increasingly generate non-verbal vocalizations (NVVs) to enhance naturalness, reliable speaker verification (SV) becomes essential to objectively assess identity consistency across both verbal and non-verbal segments. Yet current SV systems generalize poorly to NVVs, and fine-tuning on NVV data causes catastrophic forgetting of speech performance. We present the first systematic study across 10 NVV types and propose a framework combining frozen Data2Vec self-supervised features with ECAPA-TDNN, enhanced by a Mixture of Experts (MoE) module with learned domain-aware routing. A conditional distillation loss on speech inputs via a pretrained teacher retains speech-to-speech accuracy, while a contrastive loss bridges the speech-NVV domain gap. Our method reduces speech-NVV EER from 38.93% to 22.66% over a pretrained baseline, and improves speech EER from 13.17% to 9.24% via distillation.

Speaker-decoupled speech codecs can reduce bitrate by separating global speaker attributes from local content and prosody, while supporting voice conversion. Existing speaker-decoupled codecs face a trade-off: methods that explicitly suppress speaker leakage often rely on multi-stage or auxiliary training, whereas simpler designs can leave residual speaker information in local tokens. We propose SDP-Codec, a speaker-decoupled, pitch-injected codec trained with a single-stage optimization pipeline. SDP-Codec derives local tokens from continuous pre-quantization features of a pretrained self-supervised encoder and injects normalized F0 via a pitch encoder-decoder with global-conditioned denormalization and soft-label pitch reconstruction objective. Across 16 kHz and 24 kHz settings, SDP-Codec achieves competitive reconstruction and strong zero-shot voice conversion at comparable bitrates, with the lowest speaker-probing accuracy among compared systems, suggesting reduced speaker leakage.

Existing mean opinion score (MOS) prediction models typically predict utterance-level naturalness MOS and can be insensitive to localized pitch-accent errors. We propose Pitch-Accent-focused Speech Quality Assessment (PASQA), which explicitly targets pitch-accent correctness. To train our model, we construct a controlled Japanese accent-error dataset by changing accent patterns using an accent-controllable text-to-speech system, and compute a pseudo accent-quality score from the accent-error rate. PASQA builds on self-supervised representations and employs mora-conditioned fusion, ranking loss, an auxiliary accent-error localization task, and speaker-invariant training. Experiments show that conventional models fail to preserve the ordering by accent-error severity, whereas PASQA achieves high ordering accuracy on both seen and unseen speakers. Further, PASQA shows stronger agreement with human accent-correctness judgments. The code is available at https://github.com/lycorp-jp/PASQA.

User-defined keyword spotting (UD-KWS) enables zero-shot wake-word detection from text, but existing systems learn speaker-invariant representations that cannot reject impostors uttering the correct keyword. We address this dual zero-shot setting -- unseen keywords and unseen speakers -- with ZP-KWS, a lightweight framework combining a phoneme-supervised audio encoder with a GE2E-pretrained compact speaker encoder (about 0.9M parameters). Multiplicative late fusion at inference grants each branch independent veto power, supporting modes from conventional detection to strict speaker-gated activation without retraining. On LibriPhrase, Google Speech Commands, and Qualcomm datasets, ZP-KWS reduces target-only FRR at 1% FAR by up to 60% relative to the strongest baseline while maintaining competitive keyword detection, all within a 1.55M parameter budget for edge deployment.

Mean opinion score (MOS) prediction models are widely used as proxy metrics in text-to-speech (TTS) research, yet their ability to capture quality differences beyond acoustic fidelity remains unclear. We investigate this via controlled perturbations on speech: acoustic degradation, prosodic errors, and manipulation of speaker-specific characteristics such as pitch and speaking rate. We obtained MOS predictions for these speech samples from both human listeners and the model, and analyzed the differences in their perceptual characteristics. Results show that most models track acoustic degradation well, while all are insensitive to prosodic errors despite large subjective score drops. For speaker characteristics, models exhibit a double dissociation: strong mean fundamental frequency (F0) biases absent in human ratings, yet insensitivity to speaking rate and F0 variability that humans notice. These findings highlight limitations of scalar MOS prediction beyond acoustic fidelity.

Voice attribute editing models modify characteristics such as age and gender while preserving speaker identity. In large-scale speech datasets, however, attribute annotations are often noisy or inconsistent, which can cause conditional generative models to produce unstable edits. In this work, we show that idempotency provides an effective mechanism for improving robustness to noisy labels. An idempotent operator is one for which repeated application does not change the result, i.e., f(f(x)) = f(x). Enforcing this property acts as an implicit regularizer that reduces sensitivity to mislabeled examples. We introduce RIVET, a training framework that incorporates an idempotency objective to improve robustness to label noise. We evaluate RIVET under controlled label noise and on the GLOBE dataset with naturally noisy annotations. RIVET improves editing success and better preserves speaker identity than standard training, showing that idempotency improves robustness in voice editing models.

Mean opinion scores (MOS) are widely used for speech quality assessment, yet scalar labels are sensitive to rater variability and listening test differences. This introduces labeling noise, which limits the reliability of MOS prediction. Preference prediction reduces this variability as listeners compare signals directly, producing cleaner labels. We study MOS-free preference prediction and propose PrefSQA, which incorporates uncertainty-aware logits, an impairment attention head, and a module based on non-matching-reference comparisons. We use and refine five datasets, including MOS-derived and low-noise simulated sets with matching and non-matching content, experiment with human preference sets, and test on unseen data. Experiments show small improvements on MOS-derived data, while other sets reveal clear improvement over the baselines, highlighting the value of high-quality preference data and demonstrating the effectiveness of the proposed method.

Early detection of cognitive impairment relies on neuropsychological tests to minimize subjectivity by assessing multiple cognitive domains. Speech-based evaluation can support diagnostics and improve accessibility, but transcription errors and the omission of nonverbal subtests (e.g., motor skills) limit accuracy. Beyond conventional test scores, speech-derived features can provide additional insights into cognitive status. This study investigates the speech-based evaluation of the German "Syndrom-Kurz-Test," a standardized dementia screening test comprising verbal and motor subtests. We train models that integrate transcript-derived scores and Whisper embeddings per verbal subtest to reduce scoring errors. To compensate for missing motor subtests, we then leverage these fused representations to approximate expert overall ratings. Despite omitting subtests, our models strongly correlate with expert ratings and efficiently and accurately discriminate between cognitive status groups.

Dysarthria is a speech disorder marked by reduced intelligibility and communicative effectiveness. Automatic utterance-level assessment of dysarthric speech can support scalable speech monitoring and therapy-related analysis. Yet training such systems is bottlenecked by the scarcity of clinically annotated dysarthric speech. This work proposes to augment dysarthric speech assessment using data from speech synthesis evaluations, specifically human-annotated utterances with Mean Opinion Score (MOS) labels from the QualiSpeech corpus. Experiments show that fine-tuning on speech synthesis assessment data consistently improves performance on both intelligibility and naturalness prediction, while joint training yields gains primarily on naturalness. These results suggest that synthesis artifacts and dysarthric speech share perceptual commonalities, and speech synthesis evaluation corpora offer a practical augmentation source that reduces reliance on scarce clinical annotations.

Speech foundation models, pre-trained on large corpora of unlabelled speech data, produce general-purpose representations which are useful across tasks. However, these representations encode information about salient speech variables in a distributed manner, while downstream speech tasks rely on only some of this variability. In this work, we propose a post-training refinement approach using interventional contrastive learning. By leveraging an interventional dataset and multi-part contrastive loss, we learn a transformation from the entangled representation space of speech foundation models into separate content and speaker subspaces. We evaluate the learnt representations on speaker verification and keyword spotting tasks, showing improved out-of-domain speaker verification performance and evidence that speaker and content information are separated across the learned subspaces.

Multilingual speaker verification remains challenging because language-dependent acoustic variability causes speaker identity to become entangled with linguistic characteristics, degrading generalization across languages. In multilingual training, embeddings often encode language cues with speaker identity, causing speakers to form language-specific clusters. We propose L-Proto, a language-aware episodic prototypical training strategy that constructs language-consistent episodes. By sampling speakers from a single language per episode, L-Proto reduces language-driven variation during training and encourages embeddings to focus more directly on speaker identity. Experiments on the TidyVoice Challenge benchmark demonstrate consistent performance improvements over conventional fine-tuning and random episodic sampling across multiple backbone architectures.

It is well known that intelligibility of speech reduces in the presence of ambient noise. However, studies show that all sounds are not affected uniformly (or equally) and that vowels are more robust to noise than consonants. In this study, intelligibility of various consonants is assessed and analyzed in stationary white noise and non-stationary babble noise conditions. Specifically, this study investigates the individual contribution of magnitude and phase spectra of a given speech signal on human speech recognition of consonants in noisy conditions. In this regard, three experiments are carried out. In experiment 1, clean signal, signal reconstructed with only magnitude spectrum information (magnitude only signal) and signal reconstructed with only phase spectrum information (phase only signal) are assessed for intelligibility. In experiment 2, noise is added to clean speech. From noisy speech, phase only signal and magnitude only signal are reconstructed and intelligibility tests are performed for all these three signals. In experiment 3, noise is added directly to the magnitude only and phase only signals reconstructed from clean speech and their intelligibility is assessed. Results of these experiments show that magnitude spectrum contributes more to intelligibility in clean condition than phase spectrum, while information from phase spectrum is more robust in noisy conditions. It is also observed that, among consonants, nasals are more susceptible to noise whereas fricatives and approximants were observed to be comparatively more robust.

Understanding speaker confidence is crucial in educational settings, as it can enhance personalised feedback and improve learning outcomes. This study introduces a novel framework for detecting speaker confidence by integrating human-engineered features with embeddings from the Whisper encoder. To address data limitations, a pseudo-labelling technique is employed to expand the labelled dataset, allowing the model to learn from both human-annotated and model-generated labels. The framework combines traditional speech features including pitch, volume, rate of speech, and the presence of disfluencies and stress, with Whisper embeddings, and uses a co-attention mechanism to fuse these representations and achieve an overall accuracy of 75%. This study contributes to advancing speech analysis, enabling applications that support personalised learning and speaking skill development.

Short-duration speaker verification (SDSV) is crucial for personalized keyword spotting, where test utterances are typically shorter than three seconds. Limited speech duration results in unstable speaker representations and increased sensitivity to noise and phoneme variations, thereby degrading performance. To investigate this issue, we construct VoxPhrase, a large-scale SDSV corpus automatically segmented from the VoxCeleb dataset. Our analysis shows that text-dependent (TD) enrollment is constrained by duration and yields unstable speaker representations. In contrast, although text-independent (TI) enrollment introduces content mismatch, its representations become more stable as the enrollment duration increases. Accordingly, we propose a hybrid-enrollment neural re-scoring framework that combines TD and TI enrollment and performs frame-level comparison via parallel cross-attention. Experiments on VoxPhrase demonstrate consistent improvements across multiple speaker models.

Non-verbal vocalizations (NVs), such as laughter, sighs, and coughs, are important acoustic cues for emotion and intent. Existing speech quality assessment methods typically focus on overall naturalness, while non-verbal TTS evaluations mainly examine whether a target NV appears with the correct type and position. However, the perceptual quality of NV events themselves remains underexplored. To address this gap, we construct an NV-MOS dataset containing outputs from multiple NV-TTS systems and naturally occurring NV samples, with ratings collected from three acoustic experts on a perceptual quality scale. We further analyze audio-capable multimodal large language models such as Gemini and find clear inconsistencies between their scores and expert ratings. These results suggest that general-purpose multimodal models cannot reliably replace human judgments for NV quality assessment. We then propose NVMOS, to our knowledge the first model that can reliably predict the perceptual quality of NV events in speech. Experimental results show that, with a local NV-event focusing module, NVMOS reaches expert-level or stronger agreement with human MOS.

A key challenge of speaker de-identification is the balance between privacy and utility. Many utility variables, such as the cognitive health status of the speaker, are correlated with the privacy variable, such as the speaker identity, violating the independence assumption held by the disentanglement-based approaches, causing leakage of private information and the loss of useful information for downstream tasks. To tackle this challenge, we propose a general framework, DDPO-VC, for speaker de-identification through reinforcement learning-based post-training with diffusion models. Learning from reward signals combining knowledge from privacy-focused and utility-focused teachers, our method outperforms various strong \deid/ methods in both privacy preservation and cognitive utility on two commonly used dementia speech benchmarks. Please check out our code\footnote{\href{https://github.com/cactuswiththoughts/DDPO-VC}{https://github.com/cactuswiththoughts/DDPO-VC}} and demo\footnote{\href{https://cactuswiththoughts.github.io/SpeakerDeID-Demo/}{https://cactuswiththoughts.github.io/SpeakerDeID-Demo/}}.

Speech recognition is challenging for dysarthric speakers. While federated learning (FL)-based ASR can be an effective tool for protecting privacy, it suffers from heterogeneity issues caused by speaker variability. Forcing all speakers to share the same model components can be suboptimal under such heterogeneity, making personalization a promising direction; however, related research on dysarthric speech remains limited. To this end, this paper explores two aggregation strategies to achieve personalization, including the parameter-based averaging strategy and the embedding-based averaging strategy. Experiments on UASpeech and TORGO show that the proposed methods outperform the baseline regularized FedAvg by statistically significant WER reductions of up to 0.99% absolute (3.15% relative) on UASpeech and 0.56% absolute (4.73% relative) on TORGO, respectively.

Open-set source tracing is increasingly framed as a verification problem, motivating the use of pairwise metric-learning objectives from biometrics. We thus compare global anchoring and pairwise verification under matched backbones and a fixed data and epoch budget on MLAAD (in-domain) and STOPA (out-of-domain). In our runs, global anchoring yields lower in-domain error (8.61% EER) than pairwise variants (12-15% EER), even with rival mining and XLS-R finetuning. Because pairwise objectives optimize similarity directly, they concentrate variance into fewer embedding directions, reducing resolution among closely related generators. To test if this drives the drop, we impose a similar bottleneck to the globally supervised baseline, yet the baseline remains competitive. Together with an embedding-space analysis ($k_{99}$), these results suggest that the gap is not explained by dimensionality alone, but rather by the pairwise objective's shaping of the retained directions.

Spoken factual claims often occur within multi-turn conversations, where surrounding dialogue can provide context unavailable from the claim alone. Yet most fact-checking research evaluates isolated text, leaving conversational audio under-studied. We introduce MAD2, a synthetic Multi-turn Audio Dialogues benchmark for spoken claim verification with 1,000 two-speaker dialogues, 1,230 sentence-level check-worthy candidate annotations, and approximately 10 hours of audio. We also propose calibrated multimodal fusion of a context-aware audio encoder and a dialogue-aware text model. Adding dialogue context improves verification across settings, although the gains vary by scenario. Past-only context often approaches local offline performance, suggesting its usefulness when future context is unavailable. Fusion achieves its highest mean advantage over text with full-dialogue context, but does not consistently or significantly outperform text across settings. Under full-dialogue context, exploratory subgroup results show greater performance variation across dialogue scenarios for text and fusion, but across spread styles for audio.

We introduce a spoofing countermeasure architecture conditioned on speaker-reference recordings, but observe that it converges to a solution that effectively ignores the reference during inference. Surprisingly, training with a reference channel induces invariance that improves deepfake detection, even when the reference is absent or mismatched during inference. Based on this observation, we propose a Reference-Augmented Training (RAT) strategy. RAT yields improved detection performance compared to single-utterance baselines, even when the reference recording is replaced with a zero vector at inference. Through rigorous analysis, we demonstrate that the optimization process rapidly diminishes the reference contributions, leading to inference largely independent of the reference channel. Using RAT, we achieve state-of-the-art 2.57% EER and 0.074 minDCF on the ASVspoof 5 benchmark with a single detector, surpassing even large ensemble systems.

The Environment-Aware Speech and Sound Deepfake Detection Challenge (ESDD2), held in conjunction with ICME 2026, evaluated systems for five component-level audio spoofing detection, where speech and environmental sounds may be manipulated independently or jointly. After the challenge concludes, we analyze the final leaderboard and summarize effective design choices from the top-performing submissions. The challenge attracted 94 registrations from 16 countries; after verification of submission requirements and metadata, 13 teams were retained for the final analysis. On the test set, the best system achieved a Macro-F1 score of 0.8775, substantially outperforming the separation-enhanced joint learning baseline (0.6327). Top systems consistently benefited from modular task decomposition, cross-domain self-supervised encoders, targeted data augmentation, and selective ensembling rather than simple model scaling. At the same time, auxiliary EER analyses reveal persistent difficulty in detecting the spoofed environmental component and in generalizing to unseen generators in the test set. This paper reports challenge results and provides insights for future environment-aware deepfake detection research. The CompSpoofV2 dataset and baseline code remain publicly available for reproducibility.

Neural speech codecs enable low-bitrate speech communication, yet at ultra-low bitrates (< 1000 bps) preserving perceptual quality and intelligibility is challenging. Existing designs often prioritize acoustic details, leaving limited capacity for the core linguistic message under tight bitrate constraints. To address this, we propose ContextCodec, a codec that transmits content-focused context features to explicitly guide reconstruction. ContextCodec adopts a dual-branch encoder that decouples acoustic details from content-focused context. The context branch is trained with a CLIP-style contrastive loss that aligns context features with phoneme indices, reducing paralinguistic leakage. During decoding, these features are injected at each decoding stage for explicit guidance. In addition, we introduce a lightweight autoregressive latent refinement module. Experiments show a strong quality-intelligibility trade-off down to 500 bps, with an RTF of 0.4886 on a typical mobile CPU.

This paper presents a lightweight, cascaded GMM-DTW dual-factor voice lock system for resource-constrained edge environments. By utilizing a shared MFCC feature space, the framework implements a sequential defense mechanism combining GMM speaker screening and DTW passphrase verification. To counter presentation threats without extra hardware, a dynamic joint absolute-relative margin constraint is integrated into the GMM classification space, limiting the physical imposter and high-fidelity replay attack False Acceptance Rates (FAR) to 2.73% and 6.67%, respectively, with a legitimate False Rejection Rate (FRR) of 16.67%. Due to Sakoe-Chiba window optimization, the global end-to-end processing latency under temporal stress is rigidly bounded at 9.82ms on a single-core CPU, comprising 1.51ms for feature extraction, 0.54ms for GMM scoring, and 7.77ms for worst-case DTW matching. These empirical benchmarks demonstrate the viability of white-box acoustic cascades for secure, deterministic real-time deployment on low-power edge nodes.

While speech quality is typically assessed on complete utterances, streaming and generative systems require incremental estimation from partial audio. Existing predictors assume full context, degrading on prefix-constrained inputs. Extending ARECHO, we propose ANCHOR, reformulating incremental assessment as a multi-resolution autoregressive task. It models chunk- and utterance-level quality within a single decoder using dual-resolution tokens and a resolution-aware hierarchy for coarse-to-fine refinement. Experiments show substantial robustness under partial input, including a 48% PLCMOS error reduction on 2-second prefixes. Convergence analysis reveals a 4-6 s effective perceptual context horizon. A stress test further isolates structured extrapolation biases under localized corruption. Results demonstrate that hierarchical supervision improves incremental prediction and elucidates how perceptual quality accumulates over time.

We present a voice conversion (VC) framework that utilizes K-Nearest Neighbors (KNN) retrieval over WavLM representations to align non-parallel source and target speech, constructing synthetic training pairs for supervised learning. The retrieved segments serve as synthetic inputs, while real target audio provides ground-truth outputs, forming a synthetic-to-real training paradigm that naturally supports multilingual data without requiring parallel corpora or explicit alignment. To ensure consistent target-speaker identity, we incorporate a speaker loss derived from a pretrained speaker verification model. Experiments across multiple languages demonstrate that the proposed approach achieves high naturalness and strong speaker similarity, outperforming competitive VC baselines, despite being trained exclusively on English data. Samples can be accessed at: https://palindromic-vc.github.io.

Sophisticated generative speech technology can undermined the reliability of voice biometrics. While spoofing detection systems excel when assessed under in-domain conditions, generalisation to out-of-domain settings is often poor. In this paper, we show that such issues could be caused by speaker bias, where models learn individual voice traits rather than markers of manipulation or generation. We propose a teacher-student framework for speaker-invariant spoofing detection that disentangles identity without requiring speaker labels. We leverage a pre-trained speaker recognition teacher to guide a student model via a gradient reversal layer. To control the balance between suppressing cues related to voice identity with the preservation of those related to spoofing detection, we integrate a Variational Information Bottleneck. Evaluations across nine datasets show our model achieves a 25.7% relative reduction to the EER compared to the MHFA baseline.

Voice biometric systems face growing threats from spoofing attacks, yet the evaluation of detection models remains inconsistent across datasets. To investigate these unpredictable fluctuations, we conduct a comprehensive benchmark of four self-supervised learning feature extractors paired with four back-end classifiers. We compare the hierarchical local feature extraction of ResNet with the global sequence and relational modeling of attention and graph-based back-ends. Through multi-corpus training across three scenarios and six evaluation datasets, our empirical analysis yields two critical findings. First, we expose a domain bias within the ASVspoof 5 dataset, showing that naive data scaling actively degrades performance. Second, our cross-linguistic analysis reveals that fine-tuning with just 8 hours of target-language data enhances detection robustness. Together, these findings emphasize the critical need for domain-aware and language-specific adaptation in spoofing detection.

Deep-learning speaker verification (SV) increasingly relies on deep neural network backbones, whose environmental impact remains largely undocumented. In this paper, we conduct an evaluation of ResNet architectures trained on VoxCeleb2, varying depth, channel width, and stage distribution, and measure energy consumption and carbon footprint using node-level sensors. Results show a clear point of diminishing returns: deeper or wider models bring only marginal accuracy gains while energy consumption grows steeply. In contrast, mid-sized networks such as ResNet-50 and stage-concentrated variants achieve favorable trade-offs between performance and environmental impact. These findings provide actionable guidelines for designing energy-efficient SV systems.

Although low-bit quantization provides practical means to deploy speaker verification on resource-constrained devices, its effects on speaker verification performance remain poorly understood. In this paper, we study uniform K-means quantization-aware training of ResNet-36 and ResNet-200 through joint layer-wise and score-level analyses. Our layer-wise analysis highlights fragile components and shows that score degradation is not fully explained by weight distortion alone. We identify a clear knee point at 2 bits, with larger score drift and harmful decision flips concentrated near the FP32 threshold. Our score-level analysis reveals where and how score errors emerge under extreme quantization. Building on these findings, we propose a calibrated multi-precision cascade that resolves most trials at 2 bits and escalates only ambiguous cases, achieving performance close to FP32 while preserving the efficiency benefits of low-bit inference with substantially lower compute and memory costs.

Cross-lingual voice cloning aims to generate speech in a target language while preserving speaker identity from a source-language reference. This task is central to speech translation and is the focus of the IWSLT 2026 Cross-Lingual Voice Cloning track. A key challenge is maintaining intelligibility and naturalness in the presence of accent variation and domain-specific vocabulary. We build on a multilingual text-to-speech model, FishAudio-S2-Pro, and introduce language tag prompting to improve language control and reduce accent leakage. We further apply reinforcement learning (RL) fine-tuning for task adaptation and observe improvements in intelligibility. Finally, we propose a reference-conditioned lexical matching method that improves pronunciation of domain-specific terms when lexical overlap is present. Results show that language prompting provides the largest gains, while lexical matching yields consistent improvements on matched subsets.

Speech anonymization is commonly evaluated using averagecase metrics such as the equal error rate, which can hide large disparities in re-identification risks across individuals. In this paper, we conduct a large-scale per-speaker privacy analysis using a linkability-based metric under a worst-case scenario. Nearly 5,000 speakers are evaluated across multiple anonymization systems, attacker architectures, and conversation lengths. While linkability scores are highly polarized at the speaker level, the sets of easy to re-identify and hard to re-identify speakers vary substantially across configurations. We show that no single factor explains speaker vulnerability. Instead, the re-identification risk emerges from the interaction between the attacker, the anonymizer, and the amount of available speech. These results challenge the notion of intrinsic speaker-level privacy risks and emphasize the need for evaluation protocols that are explicitly conditioned on the attacker and anonymizer.

Scripted vs spontaneous speech detection is appealing for interview guardrails, but benchmark performance can be inflated by shortcuts tied to corpus identity, channel conditions, and recording artifacts rather than speaking style itself. We present SEAM, a shortcut-aware framework for real-time scriptedness detection that combines uniform preprocessing, seam-aware sampling, non-speech augmentation, and a compact DistilHuBERT backbone. With 8s windows, the model achieves 0.971 +- 0.004 ROC-AUC on an external interview-domain evaluation set. Removing the shortcut-prevention components improves internal held-out metrics but sharply reduces external performance, indicating shortcut learning. Post-training quantization reduces the model footprint to 41.8MB with little loss in external performance. The results demonstrate that robust real-time scriptedness detection depends not only on the backbone, but on shortcut-aware data design and evaluation. We release code and model checkpoints.

Building a lexicon from discovered word-like units is a central goal in zero-resource speech processing. But do our evaluations provide a trustworthy indication of lexicon quality? A common metric, normalized edit distance, averages the phoneme edit distances between discovered units in each cluster. We show that this metric has an inherent bias toward the quality of large clusters, inhibiting fair evaluation. Moreover, it ignores how well true classes are distributed across clusters. Based on established theory in clustering literature, we propose two metrics that address these shortcomings: a modified metric that weighs cluster size when assessing within-cluster consistency, and an inverse metric that assesses how true words are spread across clusters. Through experiments on synthetic and real-world lexicons, we demonstrate that combined, these metrics are: (1) more closely correlated with how similar a lexicon is to the ground-truth distribution, and (2) more robust to biases that skew lexicon evaluations.

We present ProSarc, an audio-only framework that detects sarcasm by modelling temporal prosodic incongruity, that is, the mismatch between local prosodic dynamics and the utterance-level emotional baseline. Dual encoding paths, a Global Emotion Encoder and a Temporal Prosody Encoder (BiLSTM + multi-head attention), feed a Prosodic Incongruity Analyzer that produces a scalar incongruity score for classification. Monte Carlo dropout provides uncertainty estimates, and an attention-based mechanism localises sarcastic onset without frame-level labels. ProSarc outperforms prior audio-only methods on MUStARD++ (F1=75.3) and generalises to spontaneous (PodSarc, F1=62.9) and cross-lingual speech (MuSaG, F1=65.6). Ten-run validation confirms the contribution of incongruity modelling (Wilcoxon p=0.002, Cohen's d=1.51). Human evaluation shows that model uncertainty tracks perceptual ambiguity and predicted onsets align with human-annotated temporal windows.

We propose GLASS, a framework for composable acoustic style control in zero-shot autoregressive text-to-speech (TTS) that learns controls from post-generation rewards rather than style labels. In zero-shot TTS, a speaker prompt often entangles speaker identity with prosodic attributes such as speaking rate and pitch, making it difficult to change style without changing the prompt itself. GLASS instead treats each acoustic attribute as a reward-defined control direction. For each control axis, GLASS freezes the TTS backbone and trains one lightweight LoRA adapter with Group Relative Policy Optimization (GRPO), using speech-token length and mean F0 as style rewards and WER as an intelligibility anchor. Because each control is represented as a LoRA weight update, independently trained adapters can be swapped, interpolated, and composed through linear LoRA arithmetic without retraining the backbone. Experiments on speaking rate and pitch control show targeted style shifts while preserving naturalness, speaker similarity, and intelligibility, and demonstrate smooth interpolation and multi-axis composition across independently trained adapters.