Speech_Recognition
-
Multimodal LLMs are not all you need for Pediatric Speech Language Pathology
2026-04-29 11:52:54Darren F\"urst, Sebastian Steindl, Ulrich Sch\"aferAbstract
Speech Sound Disorders (SSD) affect roughly five percent of children, yet speech-language pathologists face severe staffing shortages and unmanageable caseloads. We test a hierarchical approach to SSD classification on the granular multi-task SLPHelmUltraSuitePlus benchmark. We propose a cascading approach from binary classification to type, and symptom classification. By fine-tuning Speech Representation Models (SRM), and using targeted data augmentation we mitigate biases found by previous works, and improve upon all clinical tasks in the benchmark. We also treat Automatic Speech Recognition (ASR) with our data augmentation approach. Our results demonstrate that SRM consistently outperform the LLM-based state-of-the-art across all evaluated tasks by a large margin. We publish our models and code to foster future research.
Abstract (translated)
URL
https://arxiv.org/abs/2604.26568
PDF
-
Text-Utilization for Encoder-dominated Speech Recognition Models
2026-04-29 10:28:00Albert Zeyer, Tim Posielek, Ralf Schl\"uter, Hermann NeyAbstract
This paper investigates efficient methods for utilizing text-only data to improve speech recognition, focusing on encoder-dominated models that facilitate faster recognition. We provide a comprehensive comparison of techniques to integrate text-only data, including modality matching and dynamic downsampling to reach text-level representations within the encoder. Our experiments on the LibriSpeech corpus show that a larger encoder with a smaller decoder can equal or surpass the performance of architectures with larger decoders. We demonstrate that simple configurations, such as random duration models, are often more effective than complex alternatives, significantly simplifying the training pipeline. All code and recipes are made publicly available.
Abstract (translated)
URL
https://arxiv.org/abs/2604.26514
PDF
-
WhisperPipe: A Resource-Efficient Streaming Architecture for Real-Time Automatic Speech Recognition
2026-04-28 13:18:59Erfan Ramezani, Mohammad Mahdi Giahi, Mohammad Erfan Zarabadipour, Amir Reza Yosefian, Hamid GhadiriAbstract
Real-time automatic speech recognition (ASR) systems face a fundamental trade-off between transcription accuracy and computational efficiency, particularly when deploying large-scale transformer models like Whisper. Existing streaming approaches either sacrifice accuracy through aggressive chunking or incur prohibitive memory costs through unbounded context accumulation. We present WhisperPipe, a novel streaming architecture that achieves bounded memory consumption while maintaining transcription quality through three key innovations a hybrid Voice Activity Detection (VAD) pipeline combining Silero VAD with energy-based filtering to reduce false activations by 34%, a dynamic buffering mechanism with overlapping context windows that prevents information loss at segment boundaries, and an adaptive processing strategy that balances latency and accuracy based on speech characteristics. Evaluated on 2.5 hours of diverse audio data, WhisperPipe demonstrates a median end-to-end latency of 89ms (90th percentile: 142ms) while consuming 48% less peak GPU memory and 80.9% lower average GPU utilization compared to baseline Whisper implementations. The system maintains stable memory usage over extended sessions, with zero growth rate across 150-minute continuous operation. Comparative analysis against related work shows that WhisperPipe achieves competitive accuracy (WER within 2% of offline Whisper) while operating at 3-5x lower latency than existing streaming solutions. The architecture's modular design enables deployment across resource-constrained environments, from edge devices to cloud infrastructure. Our results demonstrate that careful architectural design can reconcile the competing demands of real-time responsiveness and model sophistication in production ASR systems.
Abstract (translated)
URL
https://arxiv.org/abs/2604.25611
PDF
-
RAS: a Reliability Oriented Metric for Automatic Speech Recognition
2026-04-27 10:11:08Wenbin Huang, Yuhang Qiu, Bohan Li, Yiwei Guo, Jing Peng, Hankun Wang, Xie Chen, Kai YuAbstract
Automatic speech recognition systems often produce confident yet incorrect transcriptions under noisy or ambiguous conditions, which can be misleading for both users and downstream applications. Standard evaluation based on Word Error Rate focuses solely on accuracy and fails to capture transcription reliability. We introduce an abstention-aware transcription framework that enables ASR models to explicitly abstain from uncertain segments. To evaluate reliability under abstention, we propose RAS, a reliability-oriented metric that balances transcription informativeness and error aversion, with its trade-off parameter calibrated by human preference. We then train an abstention-aware ASR model through supervised bootstrapping followed by reinforcement learning. Our experiments demonstrate substantial improvements in transcription reliability while maintaining competitive accuracy.
Abstract (translated)
URL
https://arxiv.org/abs/2604.24278
PDF
-
2nd of the 5th PVUW MeViS-Audio Track: ASR-SaSaSa2VA
2026-04-27 01:19:56Zhiyu Wang, Xudong Kang, Shutao LiAbstract
Audio-based video object segmentation aims to locate and segment objects in videos conditioned on audio cues, requiring precise understanding of both appearance and motion. Recent audio-driven video segmentation methods extend MLLMs by fusing audio and visual features for end-to-end localization. Despite their promise, these approaches are computationally intensive, struggle with aligning temporal audio cues to dynamic video content, and depend on large paired audio-video datasets. To address these challenges, we present ASR-SaSaSa2VA, a resource-efficient framework for audio-guided video segmentation. The key idea is to convert audio inputs into textual motion descriptions via automatic speech recognition (ASR) models and then leverage pre-trained text-based referring video segmentation models (e.g., SaSaSa2VA) for pixel-level predictions. To further enhance robustness, we incorporate a no-target expression detection module, implemented by a fine-tuned audio-based MLLM, which filters out audio clips that do not refer to any target object. This design allows the system to exploit strong pre-trained models while effectively handling ambiguous or irrelevant audio inputs. Our approach achieves a final score of 80.7 in the 5th PVUW Challenge (MeViS-v2-Audio track), earning the second-place ranking.
Abstract (translated)
URL
https://arxiv.org/abs/2604.23935
PDF
-
Keyword spotting using convolutional neural network for speech recognition in Hindi
2026-04-26 21:11:29Saru Bharti, Pushparaj Mani PathakAbstract
In this study, we investigate the application of keyword spotting (KWS) in the domain of Hindi speech recognition, utilizing a dataset comprising 40,000 audio samples. With a sampling rate of 44 kHz and an average duration of 1.9 seconds per sample, we focus on developing an efficient on-device KWS system tailored for user-specific queries. Leveraging Convolutional Neural Networks (CNNs) for classification, we employ feature engineering techniques to convert raw audio recordings into Mel Frequency Cepstral Coefficients (MFCCs) as an input for our network. Our experiments encompass various CNN architectures, exploring their efficacy in identifying predefined keywords within the continuous speech stream. Our CNN-based approach achieves a commendable accuracy rate of 91.79% through rigorous evaluation, demonstrating promising performance while ensuring computational efficiency and user-specific customization in Hindi speech recognition.
Abstract (translated)
URL
https://arxiv.org/abs/2605.02928
PDF
-
Au-M-ol: A Unified Model for Medical Audio and Language Understanding
2026-04-25 12:57:25Meizhu Liu, Nistha Mitra, Paul Li, Amine Abdaoui, Adam Ledyard, Tao ShengAbstract
In this work, we present Au-M-ol, a novel multimodal architecture that extends Large Language Models (LLMs) with audio processing. It is designed to improve performance on clinically relevant tasks such as Automatic Speech Recognition (ASR). Au-M-ol has three main components: (1) an audio encoder that extracts rich acoustic features from medical speech, (2) an adaptation layer that maps audio features into the LLM input space, and (3) a pretrained LLM that performs transcription and clinical language understanding. This design allows the model to interpret spoken medical content directly, improving both accuracy and robustness. In experiments, Au-M-ol reduces Word Error Rate (WER) by 56\% compared to state-of-the-art baselines on medical transcription tasks. The model also performs well in challenging conditions, including noisy environments, domain-specific terminology, and speaker variability. These results suggest that Au-M-ol is a strong candidate for real-world clinical applications, where reliable and context-aware audio understanding is essential.
Abstract (translated)
URL
https://arxiv.org/abs/2604.23284
PDF
-
Identifying and typifying demographic unfairness in phoneme-level embeddings of self-supervised speech recognition models
2026-04-24 15:03:39Felix Herron, Solange Rossato, Alexandre Allauzen, Fran\c{c}ois PortetAbstract
Modern automatic speech recognition (ASR) systems have been observed to function better for certain speaker groups (SGs) than others, despite recent gains in overall performance. One potential impediment to progress towards fairer ASR is a more nuanced understanding of the types of modeling errors that speech encoder models make, and in particular the difference between the structure of embeddings for high-performance and low-performance SGs. This paper proposes a framework typifying two types of error that can occur in modeling phonemes in ASR systems: random error/high variance in phoneme embedding, vs systematic error/embedding bias. We find that training phoneme classification probes only on a single, typically disadvantaged SG, sometimes improves performance for that SG, which is evidence for the existence of SG-level bias in phoneme embeddings. On the other hand, we find that speakers and SGs with higher levels of phoneme variance are the same as those with worse phoneme prediction accuracy. We conclude that both types of error are present in phoneme embeddings and both are candidate causes for SG-level unfairness in ASR, though random error is likely a greater hindrance to fairness than systematic error. Furthermore, we find that finetuning encoder models using a fairness-enhancing algorithm (domain enhancing and adversarial training) changes neither the benefits of in-domain phoneme classification probe training, nor measured levels of random embedding error.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22631
PDF
-
Advancing automatic speech recognition using feature fusion with self-supervised learning features: A case study on Fearless Steps Apollo corpus
2026-04-24 04:15:43Szu-Jui Chen, John H. L. HansenAbstract
Using self-supervised learning (SSL) models has significantly improved performance for downstream speech tasks, surpassing the capabilities of traditional hand-crafted features. This study investigates the amalgamation of SSL models, with the aim to leverage both their individual strengths and refine extracted features to achieve improved speech recognition models for naturalistic scenarios. Our research investigates the massive naturalistic Fearless Steps (FS) APOLLO resource, with particular focus on the FS Challenge (FSC) Phase-4 corpus, providing the inaugural analysis of this dataset. Additionally, we incorporate the CHiME-6 dataset to evaluate performance across diverse naturalistic speech scenarios. While exploring previously proposed Feature Refinement Loss and fusion methods, we found these methods to be less effective on the FSC Phase-4 corpus. To address this, we introduce a novel deep cross-attention (DCA) fusion method, designed to elevate performance, especially for the FSC Phase-4 corpus. Our objective is to foster creation of superior FS APOLLO community resources, catering to the diverse needs of researchers across various disciplines. The proposed solution achieves an absolute +1.1% improvement in WER, providing effective meta-data creation for the massive FS APOLLO community resource.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22203
PDF
-
Evaluation of Automatic Speech Recognition Using Generative Large Language Models
2026-04-23 17:59:47Thibault Ba\~neras-Roux, Shashi Kumar, Driss Khalil, Sergio Burdisso, Petr Motlicek, Shiran Liu, Mickael Rouvier, Jane Wottawa, Richard DufourAbstract
Automatic Speech Recognition (ASR) is traditionally evaluated using Word Error Rate (WER), a metric that is insensitive to meaning. Embedding-based semantic metrics are better correlated with human perception, but decoder-based Large Language Models (LLMs) remain underexplored for this task. This paper evaluates their relevance through three approaches: (1) selecting the best hypothesis between two candidates, (2) computing semantic distance using generative embeddings, and (3) qualitative classification of errors. On the HATS dataset, the best LLMs achieve 92--94\% agreement with human annotators for hypothesis selection, compared to 63\% for WER, also outperforming semantic metrics. Embeddings from decoder-based LLMs show performance comparable to encoder models. Finally, LLMs offer a promising direction for interpretable and semantic ASR evaluation.
Abstract (translated)
URL
https://arxiv.org/abs/2604.21928
PDF
-
Do LLM Decoders Listen Fairly? Benchmarking How Language Model Priors Shape Bias in Speech Recognition
2026-04-23 04:40:58Srishti Ginjala, Eric Fosler-Lussier, Christopher W. Myers, Srinivasan ParthasarathyAbstract
As pretrained large language models replace task-specific decoders in speech recognition, a critical question arises: do their text-derived priors make recognition fairer or more biased across demographic groups? We evaluate nine models spanning three architectural generations (CTC with no language model, encoder-decoder with an implicit LM, and LLM-based with an explicit pretrained decoder) on about 43,000 utterances across five demographic axes (ethnicity, accent, gender, age, first language) using Common Voice 24 and Meta's Fair-Speech, a controlled-prompt dataset that eliminates vocabulary confounds. On clean audio, three findings challenge assumptions: LLM decoders do not amplify racial bias (Granite-8B has the best ethnicity fairness, max/min WER = 2.28); Whisper exhibits pathological hallucination on Indian-accented speech with a non-monotonic insertion-rate spike to 9.62% at large-v3; and audio compression predicts accent fairness more than LLM scale. We then stress-test these findings under 12 acoustic degradation conditions (noise, reverberation, silence injection, chunk masking) across both datasets, totaling 216 inference runs. Severe degradation paradoxically compresses fairness gaps as all groups converge to high WER, but silence injection amplifies Whisper's accent bias up to 4.64x by triggering demographic-selective hallucination. Under masking, Whisper enters catastrophic repetition loops (86% of 51,797 insertions) while explicit-LLM decoders produce 38x fewer insertions with near-zero repetition; high-compression audio encoding (Q-former) reintroduces repetition pathology even in LLM decoders. These results suggest that audio encoder design, not LLM scaling, is the primary lever for equitable and robust speech recognition.
Abstract (translated)
URL
https://arxiv.org/abs/2604.21276
PDF
-
'This Wasn't Made for Me': Recentering User Experience and Emotional Impact in the Evaluation of ASR Bias
2026-04-22 23:27:41Siyu Liang, Alicia Beckford WassinkAbstract
Studies on bias in Automatic Speech Recognition (ASR) tend to focus on reporting error rates for speakers of underrepresented dialects, yet less research examines the human side of system bias: how do system failures shape users' lived experiences, how do users feel about and react to them, and what emotional toll do these repeated failures exact? We conducted user experience studies across four U.S. locations (Atlanta, Gulf Coast, Miami Beach, and Tucson) representing distinct English dialect communities. Our findings reveal that most participants report technologies fail to consider their cultural backgrounds and require constant adjustment to achieve basic functionality. Despite these experiences, participants maintain high expectations for ASR performance and express strong willingness to contribute to model improvement. Qualitative analysis of open-ended narratives exposes the deeper costs of these failures. Participants report frustration, annoyance, and feelings of inadequacy, yet the emotional impact extends beyond momentary reactions. Participants recognize that systems were not designed for them, yet often internalize failures as personal inadequacy despite this critical awareness. They perform extensive invisible labor, including code-switching, hyper-articulation, and emotional management, to make failing systems functional. Meanwhile, their linguistic and cultural knowledge remains unrecognized by technologies that encode particular varieties as standard while rendering others marginal. These findings demonstrate that algorithmic fairness assessments based on accuracy metrics alone miss critical dimensions of harm: the emotional labor of managing repeated technological rejection, the cognitive burden of constant self-monitoring, and the psychological toll of feeling inadequate in one's native language variety.
Abstract (translated)
URL
https://arxiv.org/abs/2604.21148
PDF
-
Aligning Stuttered-Speech Research with End-User Needs: Scoping Review, Survey, and Guidelines
2026-04-22 13:16:23Hawau Olamide Toyin, Mutiah Apampa, Toluwani Aremu, Humaid Alblooshi, Ana Rita Valente, Gon\c{c}alo Leal, Zhengjun Yue, Zeerak Talat, Hanan AldarmakiAbstract
Atypical speech is receiving greater attention in speech technology research, but much of this work unfolds with limited interdisciplinary dialogue. For stuttered speech in particular, it is widely recognised that current speech recognition systems fall short in practice, and current evaluation methods and research priorities are not systematically grounded in end-user experiences and needs. In this work, we analyse these gaps through 1) a scoping review of papers that deal with stuttered speech and 2) a survey of 70 stakeholders, including adults who stutter and speech-language pathologists. By analysing these two perspectives, we propose a taxonomy of stuttered-speech research, identify where current research directions diverge from the needs articulated by stakeholders, and conclude by outlining concrete guidelines and directions towards addressing the real needs of the stuttering community.
Abstract (translated)
URL
https://arxiv.org/abs/2604.20535
PDF
-
ATIR: Towards Audio-Text Interleaved Contextual Retrieval
2026-04-22 07:11:58Tong Zhao, Chenghao Zhang, Yutao Zhu, Zhicheng DouAbstract
Audio carries richer information than text, including emotion, speaker traits, and environmental context, while also enabling lower-latency processing compared to speech-to-text pipelines. However, recent multimodal information retrieval research has predominantly focused on images, largely overlooking audio, especially in the setting of interleaved audio-text contextual retrieval. In this work, we introduce the Audio-Text Interleaved contextual Retrieval (ATIR) task, where queries can alternate between audio and text modalities. We construct an ATIR benchmark by integrating several Automatic Speech Recognition (ASR), QA, and retrieval datasets, ultimately unifying four types of contextual retrieval tasks. This benchmark substantially addresses the limitations of existing audio retrieval datasets in semantic retrieval. To study this task, we evaluate several off-the-shelf retrievers and train our ATIR model based on a Multimodal Large Language Model (MLLM). We further introduce a novel token compression mechanism that is orthogonal to existing compression methods, thereby alleviating the issue of excessive audio tokens in MLLM-based ATIR models. Experimental results demonstrate that our ATIR model achieves substantial improvements over strong baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2604.20267
PDF
-
Detecting Hallucinations in SpeechLLMs at Inference Time Using Attention Maps
2026-04-21 15:18:10Jonas Waldendorf, Bashar Awwad Shiekh Hasan, Evgenii TsymbalovAbstract
Hallucinations in Speech Large Language Models (SpeechLLMs) pose significant risks, yet existing detection methods typically rely on gold-standard outputs that are costly or impractical to obtain. Moreover, hallucination detection methods developed for text-based LLMs do not directly capture audio-specific signals. We investigate four attention-derived metrics: AUDIORATIO, AUDIOCONSISTENCY, AUDIOENTROPY, and TEXTENTROPY, designed to capture pathological attention patterns associated with hallucination, and train lightweight logistic regression classifiers on these features for efficient inference-time detection. Across automatic speech recognition and speech-to-text translation tasks, evaluations on Qwen-2-Audio and Voxtral-3B show that our approach outperforms uncertainty-based and prior attention-based baselines on in-domain data, achieving improvements of up to +0.23 PR-AUC, and generalises to out-of-domain ASR settings. We further find that strong performance can be achieved with approximately 100 attention heads, improving out-of-domain generalisation compared to using all heads. While effectiveness is model-dependent and task-specific training is required, our results demonstrate that attention patterns provide a valuable tool for hallucination detection in SpeechLLMs.
Abstract (translated)
URL
https://arxiv.org/abs/2604.19565
PDF
-
UAF: A Unified Audio Front-end LLM for Full-Duplex Speech Interaction
2026-04-21 08:24:55Yadong Li, Guoxin Wu, Haiping Hou, Biye LiAbstract
Full-duplex speech interaction, as the most natural and intuitive mode of human communication, is driving artificial intelligence toward more human-like conversational systems. Traditional cascaded speech processing pipelines suffer from critical limitations, including accumulated latency, information loss, and error propagation across modules. To address these issues, recent efforts focus on the end-to-end audio large language models (LLMs) like GPT-4o, which primarily unify speech understanding and generation task. However, most of these models are inherently half-duplex, and rely on a suite of separate, task-specific front-end components, such as voice activity detection (VAD) and turn-taking detection (TD). In our development of speech assistant, we observed that optimizing the speech front-end is equally crucial as advancing the back-end unified model for achieving seamless, responsive interactions. To bridge this gap, we propose the first unified audio front-end LLM (UAF) tailored for full-duplex speech systems. Our model reformulates diverse audio front-end tasks into a single auto-regressive sequence prediction problem, including VAD, TD, speaker recognition (SR), automatic speech recognition (ASR) and question answer (QA). It takes streaming fixed-duration audio chunk (e.g., 600 ms) as input, leverages a reference audio prompt to anchor the target speaker at the beginning, and regressively generates discrete tokens encoding both semantic content and system-level state controls (e.g., interruption signals). Experiments demonstrate that our model achieves leading performance across multiple audio front-end tasks and significantly enhances response latency and interruption accuracy in real-world interaction scenarios.
Abstract (translated)
URL
https://arxiv.org/abs/2604.19221
PDF
-
Voice of India: A Large-Scale Benchmark for Real-World Speech Recognition in India
2026-04-21 07:02:01Kaushal Bhogale, Manas Dhir, Amritansh Walecha, Manmeet Kaur, Vanshika Chhabra, Aaditya Pareek, Hanuman Sidh, Sagar Jain, Bhaskar Singh, Utkarsh Singh, Tahir Javed, Shobhit Banga, Mitesh M. KhapraAbstract
Existing Indic ASR benchmarks often use scripted, clean speech and leaderboard driven evaluation that encourages dataset specific overfitting. In addition, strict single reference WER penalizes natural spelling variation in Indian languages, including non standardized spellings of code-mixed English origin words. To address these limitations, we introduce Voice of India, a closed source benchmark built from unscripted telephonic conversations covering 15 major Indian languages across 139 regional clusters. The dataset contains 306230 utterances, totaling 536 hours of speech from 36691 speakers with transcripts accounting for spelling variations. We also analyze performance geographically at the district level, revealing disparities. Finally, we provide detailed analysis across factors such as audio quality, speaking rate, gender, and device type, highlighting where current ASR systems struggle and offering insights for improving real world Indic ASR systems.
Abstract (translated)
URL
https://arxiv.org/abs/2604.19151
PDF
-
Reducing the Offline-Streaming Gap for Unified ASR Transducer with Consistency Regularization
2026-04-21 04:43:03Andrei Andrusenko, Vladimir Bataev, Lilit Grigoryan, Nune Tadevosyan, Vitaly Lavrukhin, Boris GinsburgAbstract
Unification of automatic speech recognition (ASR) systems reduces development and maintenance costs, but training a single model to perform well in both offline and low-latency streaming settings remains challenging. We present a Unified ASR framework for Transducer (RNNT) training that supports both offline and streaming decoding within a single model, using chunk-limited attention with right context and dynamic chunked convolutions. To further close the gap between offline and streaming performance, we introduce an efficient Triton implementation of mode-consistency regularization for RNNT (MCR-RNNT), which encourages agreement across training modes. Experiments show that the proposed approach improves streaming accuracy at low latency while preserving offline performance and scaling to larger model sizes and training datasets. The proposed Unified ASR framework and the English model checkpoint are open-sourced.
Abstract (translated)
URL
https://arxiv.org/abs/2604.19079
PDF
-
Where Do Self-Supervised Speech Models Become Unfair?
2026-04-20 13:27:12Felix Herron, Maja Hjuler, Solange Rossato, Alexandre Allauzen, Fran\c{c}ois PortetAbstract
Speech encoder models are known to model members of some speaker groups (SGs) better than others. However, there has been little work in establishing why this occurs on a technological level. To our knowledge, we present the first layerwise fairness analysis of pretrained self-supervised speech encoder models (S3Ms), probing each embedding layer for speaker identification (SID) automatic speech recognition (ASR). We find S3Ms produce embeddings biased against certain SGs for both tasks, starting at the very first latent layers. Furthermore, we find opposite patterns of layerwise bias for SID vs ASR for all models in our study: SID bias is minimized in layers that minimize overall SID error; on the other hand, ASR bias is maximized in layers that minimize overall ASR error. The inverse bias/error relationship for ASR is unaffected when probing S3Ms that are finetuned for ASR, suggesting SG-level bias is established during pretraining and is difficult to remove.
Abstract (translated)
URL
https://arxiv.org/abs/2604.18249
PDF
-
Hard to Be Heard: Phoneme-Level ASR Analysis of Phonologically Complex, Low-Resource Endangered Languages
2026-04-20 12:54:14V. S. D. S. Mahesh Akavarapu, Michael Daniel, Gerhard J\"agerAbstract
We present a phoneme-level analysis of automatic speech recognition (ASR) for two low-resourced and phonologically complex East Caucasian languages, Archi and Rutul, based on curated and standardized speech-transcript resources totaling approximately 50 minutes and 1 hour 20 minutes of audio, respectively. Existing recordings and transcriptions are consolidated and processed into a form suitable for ASR training and evaluation. We evaluate several state-of-the-art audio and audio-language models, including wav2vec2, Whisper, and Qwen2-Audio. For wav2vec2, we introduce a language-specific phoneme vocabulary with heuristic output-layer initialization, which yields consistent improvements and achieves performance comparable to or exceeding Whisper in these extremely low-resource settings. Beyond standard word and character error rates, we conduct a detailed phoneme-level error analysis. We find that phoneme recognition accuracy strongly correlates with training frequency, exhibiting a characteristic sigmoid-shaped learning curve. For Archi, this relationship partially breaks for Whisper, pointing to model-specific generalization effects beyond what is predicted by training frequency. Overall, our results indicate that many errors attributed to phonological complexity are better explained by data scarcity. These findings demonstrate the value of phoneme-level evaluation for understanding ASR behavior in low-resource, typologically complex languages.
Abstract (translated)
URL
https://arxiv.org/abs/2604.18204
PDF
