Speaker Normalization via Voice Conversion Reveals a Human-Machine Dissociation in Dialect Classification

Caroline Kleen, Lea Fischbach, Akbar Karimi, Lucie Flek, Alfred Lameli


Abstract
This study evaluates whether Retrieval-based Voice Conversion (RVC) can be used to normalize speaker-specific variability while preserving dialect-relevant acoustic cues, and what the response of human and machine systems to this manipulation reveals about the architecture of dialect recognition. In two perception experiments, speech samples from nine German dialect regions were presented either in their original form or after conversion to a single target speaker. We compared overall accuracy, confusion structures, item-level response distributions, and the interaction between listener origin and target dialect across conditions. Human classification remained stable under voice conversion. Accuracy did not differ between conditions, confusion matrices were highly correlated, and item-level divergences were minimal. The interaction between listener origin and target dialect—reflecting systematic regional bias—remained invariant. These findings indicate that RVC does not distort perceptually relevant dialectal cues and that human dialect recognition is robust to speaker normalization. In contrast, we evaluated a deep learning model under matched conditions: model accuracy improved significantly under RVC, while human performance remained unchanged. This dissociation reframes RVC as an experimental probe for investigating the divergence between human and machine speech processing, suggesting that this divergence is rooted in fundamentally different representational architectures.
Anthology ID:
2026.dialres-1.18
Volume:
Proceedings of the First Workshop on Dialects in NLP — A Resource Perspective
Month:
May
Year:
2026
Address:
Palma de Mallorca
Editors:
Antonis Anastasopoulos, Stella Markantonatou, Angela Ralli, Marcos Zampieri, Stavros Bompolas, Vivian Stamou
Venues:
DialRes | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
177–187
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-dialres-18
DOI:
10.63317/3sqk7nxsikhp
Bibkey:
Cite (ACL):
Caroline Kleen, Lea Fischbach, Akbar Karimi, Lucie Flek, and Alfred Lameli. 2026. Speaker Normalization via Voice Conversion Reveals a Human-Machine Dissociation in Dialect Classification. In Proceedings of the First Workshop on Dialects in NLP — A Resource Perspective, pages 177–187, Palma de Mallorca. Association for Computational Linguistics.
Cite (Informal):
Speaker Normalization via Voice Conversion Reveals a Human-Machine Dissociation in Dialect Classification (Kleen et al., DialRes 2026)
Copy Citation: