mlsanityData to 19 Oct 2026

AMECxSV: Adaptive Metadata-Driven Embedding-Fusion Calibration for X-Lingual Speaker Verification

arXiv 2607.16532 ↗First posted 17 Jul 2026eess.AScited by 0

On the map: Speech and audio language models → Speaker recognition and diarization

Abstract (arXiv)

In X-lingual automatic speaker verification (ASV), fixed front-end scores vary in reliability with language match, duration, and score source. We propose AMECxSV, an adaptive metadata-driven embedding-fusion calibration backend for metadata-available settings. AMECxSV fuses trial scores with metadata to produce calibrated target posteriors, with optional posterior-confidence abstention; metadata serve as calibration context, not speaker evidence. On a development-derived speaker-disjoint held-out split, score+metadata heads reduce equal error rate (EER) from 3.15% to 2.42% for the official TidyVoice score source and from 0.64% to 0.43% for LI-MSV; the dual-score head reaches 0.43% full-coverage EER. At 0.79 coverage, abstention yields 0.03% accepted-trial EER, not a full-coverage metric. Matched score-only, metadata-permutation, and metadata-only controls support a calibration-context interpretation and limit claims to metadata-available scoring.

Key facts. No citing papers in the map yet (data through 19 Oct 2026). Placed in Speaker recognition and diarization, one of 11 directions in Speech and audio language models. Builds on 23 of its 41 references that are in the map. First posted on arXiv 17 Jul 2026.

Builds on

23 of 41 references.

Closest ancestors · 11

Same direction, newest first

  1. 2026Language-Invariant Multilingual Speaker Verification for the TidyVoice 2026 Challenge9 Mar
  2. 2026TidyVoice 2026 Challenge Evaluation Plan29 Jan
  3. 2026TidyVoice: A Curated Multilingual Dataset for Speaker Verification Derived from Common Voice22 Jan
  4. 2024ERes2NetV2: Boosting Short-Duration Speaker Verification Performance with Computational Efficiency4 Jun
  5. 20243D-Speaker-Toolkit: An Open-Source Toolkit for Multimodal Speaker Verification and Diarization29 Mar
  6. 2023CAM++: A Fast and Efficient Network for Speaker Verification Using Context-Aware Masking1 Mar
  7. 2022Disentangled representation learning for multilingual speaker recognition1 Nov
  8. 2022Wespeaker: A Research and Production oriented Speaker Embedding Learning Toolkit31 Oct
  9. 2021Tackling the Score Shift in Cross-Lingual Speaker Verification by Exploiting Language Information18 Oct
  10. 2021Simple Attention Module based Speaker Verification with Iterative noisy label detection13 Oct
  11. 2021A Speaker Verification Backend with Robust Performance across Conditions2 Feb

Other references · 12

  1. 2016Layer Normalizationcited by 5,866
  2. 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 2,135
  3. 2017VoxCeleb: a large-scale speaker identification datasetcited by 985
  4. 2020ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verificationcited by 980
  5. 2017Selective Classification for Deep Neural Networkscited by 888
  6. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 532
  7. 2021W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Trainingcited by 412
  8. 2023Seamless: Multilingual Expressive and Streaming Speech Translationcited by 235
  9. 2020The IDLAB VoxSRC-20 Submission: Large Margin Fine-Tuning and Quality-Aware Score Calibration in DNN Based Speaker Verificationcited by 69
  10. 2013The BOSARIS Toolkit: Theory, Algorithms and Code for Surviving the New DCFcited by 24
  11. 2013The PAV algorithm optimizes binary proper scoring rulescited by 9
  12. 2014A comparison of linear and non-linear calibrations for speaker recognitioncited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026A Unified Uncertainty-Aware Back-End for Speaker Verification: Scoring, Normalization, and Calibration
  2. 2026SpeakerCard-1M: An Evidence-Grounded Corpus for In-the-Wild Speaker Verification
  3. 2025Trainable Adaptive Score Normalization for Automatic Speaker Verification
  4. 2026Simple Language Normalization Wins: Cross-Lingual Speaker Verification for the TidyVoice 2026 Challenge
  5. 2026Speaker Verification with Speech-Aware LLMs: Evaluation and Augmentation
  6. 2026CoLMbo-SV: A Grounded Language Model for Explainable Speaker Verification
  7. 2024Neural Scoring: A Refreshed End-to-End Approach for Speaker Recognition in Complex Conditions
  8. 2024The SVASR System for Text-dependent Speaker Verification (TdSV) AAIC Challenge 2024