VoxKnesset: A Large-Scale Longitudinal Hebrew Speech Dataset for Aging Speaker Modeling
Authors: Yanir Marmor, Arad Zulti, David Krongauz, Adam Gabet, Yoad Snapir, Yair Lifshitz, Eran Segal
Organizations: Department of Computer Science and Applied Mathematics, Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel
Speech processing systems face a fundamental challenge: the human voice changes with age, yet few datasets support rigorous longitudinal evaluation. We introduce VoxKnesset, an open-access dataset of ~2,300 hours of Hebrew parliamentary speech spanning 2009-2025, comprising 393 speakers with recording spans of up to 15 years. Each segment includes aligned transcripts and verified demographic metadata from official parliamentary records. We benchmark modern speech embeddings (WavLM-Large, ECAPA-TDNN, Wav2Vec2-XLSR-1B) on age prediction and speaker verification under longitudinal conditions. Speaker verification EER rises from 2.15% to 4.58% over 15 years for the strongest model, and cross-sectionally trained age regressors fail to capture within-speaker aging, while longitudinally trained models recover a meaningful temporal signal. We publicly release the dataset and pipeline to support aging-robust speech systems and Hebrew speech processing.
Figures & tables
Figure 1: VoxKnesset pairs single-speaker speech segments with aligned transcripts and verified demographic labels across 16 years of parliamentary recordings.
Figure 2: Longitudinal coverage in VoxKnesset by sex. Bars show the age range per recording year; white markers indicate median age; color intensity reflects the number of unique speakers (annotated).
Figure 3: Data curation pipeline: from raw parliamentary recordings to the speaker-attributed longitudinal subset.
Figure 4: Age distributions across commonly used speech datasets, illustrating differences in demographic coverage.
Table 2: Age prediction ( R2 /MAE in years) within-domain and under leave-one-dataset-out (LODO) cross-corpus transfer, using WavLM-Large for transfer. ΔR2 measures the in-domain vs. transfer gap.
Figure 5: Longitudinal effects of speaker aging: (a) Embedding space distribution, (b) age prediction accuracy, and (c) verification performance degradation.
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.
Viet Hoang Pham, Tran Trung Nguyen, Bao Thu Ho +2
Hanoi University of Science and Technology, Hanoi, Vietnam
State-of-the-art ASR systems primarily target native adult speech, leading to substantial performance gaps for children, adolescents, and L2 speakers. We introduce AraYoungVoices, a 151.72-hour Arabic read-speech corpus from 286 speakers aged 7--18, comprising AraKids (7--12) and AraTeens (13--18). The corpus includes 146 native Arabic (L1) and 140 second-language (L2) speakers, with native speakers spanning Egyptian, Gulf, Levantine, and North African dialectal backgrounds and L2 speakers representing diverse linguistic backgrounds across the Americas, Asia, Africa, and Europe. We benchmark four pretrained ASR models under zero-shot and fine-tuned settings using unseen-speaker-&-unseen-prompt (USUP) and unseen-speaker-&-seen-prompt (USSP) evaluations. Results show that L2 speech remains substantially more challenging than L1 speech, with the largest errors observed mainly for younger L2 speakers. Age-specific fine-tuning improves the matched age group, while joint fine-tuning provides a stronger balance across populations. ASR hypotheses are also consistently closer to the standard reading prompt than to the verbatim transcription, particularly for L2 speech, suggesting partial normalization of reading deviations.
Shammur Absar Chowdhury, Zien Sheikh Ali, Houssam Eddine-Othman Lachemat +1
This paper proposes a novel cross-utterance audio-textual prompts based speaker adaptation approach for elderly speech recognition. It enables zero-shot, real-time adaptation to unseen speakers. Speech and text embeddings are extracted from the current and a few preceding utterances, before being fused in a cross-modal manner to produce compact speaker prompts that are more consistent than i/x-vectors and ECAPA-TDNN features. Experiments on the English DementiaBank Pitt and Cantonese JCCOCC MoCA elderly speech datasets suggest that the proposed online adaptation outperforms the speaker-independent (SI) model by statistically significant word error rate (WER) or character error rate (CER) reductions of 0.61% and 1.22% absolute (2.99% and 4.48% relative). Real-time factor (RTF) speed-up ratios of up to 9.83 times are obtained over offline batch-mode adaptation.
Chengxi Deng, Xurong Xie, Shujie Hu +7
The Chinese University of Hong Kong, Hong Kong SAR, China · Institute of Software, Chinese Academy of Sciences, China · National Research Council Canada, Canada