We propose a taxonomy-guided approach to semantic alignment that assigns lexicographic senses to an onomasiological taxonomy derived from the Hallig–Wartburg/Post system. Using an LLM under strict taxonomic constraints, short and heterogeneous meaning descriptions are assigned to a common conceptual space. Evaluation against expert annotation shows that run-to-run model agreement (kappa = 0.73) closely matches human agreement (kappa = 0.74), with robustness at coarse taxonomic levels and predictable degradation at finer granularity. A qualitative network analysis demonstrates the resulting potential for cross-dictionary exploration of dialectal variation in semantics.

This paper compares phonetically weighted and unweighted string distance measures in dialectometry, examining how explicit phonetic modeling affects the quantitative representation of linguistic similarity. Using narrow IPA transcriptions from the German REDE corpus, we evaluate nine measures–Levenshtein distance, bigram and trigram overlap, cosine distance, Jaro-Winkler, Jaccard similarity, the Herrgen-Schmidt measure, and the Relative Identity Value–through correlational analysis, distributional comparison, stabilization testing, and multidimensional scaling. The phonetically weighted Herrgen-Schmidt measure consistently achieves the most balanced distance dispersion, earliest stabilization, and highest linguistic plausibility. Unweighted edit-based measures reproduce the same topological structure in compressed form; distributional and overlap-based metrics introduce systematic scale distortions through exaggeration or compression. These findings establish explicit phonetic weighting as a principled and analytically efficient extension of standard dialectometric procedures. Explicit phonetic weighting enhances resolution and interpretive precision without altering the underlying relational geometry of dialect classifications.

Automatic dialect classification is commonly treated as a supervised task with a primary focus on overall accuracy. In this paper, we argue that classification errors and model uncertainty provide valuable insights into dialectal structure and variation. We analyze a speech-based dialect classification model trained on German dialect data from three generations and evaluated across 250 speaker-disjoint splits (median weighted F1=0.42). A systematic confusion analysis shows that misclassifications are largely explained by speaker diversity, dialectal similarity, geographical proximity, and speaker self-assessment. Among these factors, the number of speakers per dialect has the strongest impact on performance, while frequent confusions between closely related dialects reflect inherent linguistic similarity rather than model limitations. Generational analyses further indicate that younger speakers exhibit reduced dialectal distinctiveness, although core dialectal features remain shared across generations. By explicitly modeling classification uncertainty, the proposed approach enables the analysis of dialect transition areas and gradient dialect boundaries. Overall, this work demonstrates that automatic dialect classification can serve not only as a predictive task but also as a tool for dialectological analysis.

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.

Recent advances in speech and language technologies increasingly rely on large and diverse corpora that represent linguistic variation across dialect regions, communicative situations, and social speaker characteristics. While substantial resources are available for Standard German, comparable spoken corpora for German dialects have so far been largely lacking, limiting the development and evaluation of dialect-sensitive NLP systems. The REDE corpus addresses this gap by providing a methodologically uniform collection of spoken German for 148 locations that systematically covers all major dialect areas in Germany. It comprises contemporary recordings collected in multiple elicitation and interaction settings, capturing variation across speaking styles, situational contexts, and speaker generations. With more than 1,500 hours of speech and rich metadata on regional and social dimensions, the REDE corpus constitutes a large-scale oral resource suitable for both linguistic research and NLP applications. This paper presents the design, structure, and methodological foundations of the corpus and discusses its relevance for current speech technology requirements.

We investigate lightweight and easily applicable data augmentation techniques for dialectal audio classification. We evaluate four main methods, namely shifting pitch, interval removal, background noise insertion and interval swap as well as several subvariants on recordings from 20 German dialects. Each main method is tested across multiple hyperparameter combinations, inlcuding augmentation length, coverage ratio and number of augmentations per original sample. Our results show that frequency-based techniques, particularly frequency masking, consistently yield performance improvements, while others such as time masking or speaker-based insertion can negatively affect the results. Our comparative analysis identifies which augmentations are most effective under realistic conditions, offering simple and efficient strategies to improve dialectal speech classification.

This paper examines the effect of preprocessing techniques on spoken dialect classification using raw audio data. We focus on modifying Root Mean Square (RMS) amplitude, DC-offset, articulation rate (AR), pitch, and Harmonics-to-Noise Ratio (HNR) to assess their impact on model performance. Our analysis determines whether these features are important, irrelevant, or misleading for the classification task. To evaluate these effects, we use a pipeline that tests the significance of each acoustic feature through distortion and normalization techniques. While preprocessing did not directly improve classification accuracy, our findings reveal three key insights: deep learning models for dialect classification are generally robust to variations in the tested audio features, suggesting that normalization may not be necessary. We identify articulation rate as a critical factor, directly affecting the amount of information in audio chunks. Additionally, we demonstrate that intonation, specifically the pitch range, plays a vital role in dialect recognition.

Based on historical dialect data we introduce a local measure of linguistic coherence in spatial language variation aiming at the identification of regions which are particularly sensitive to language variation and change. Besides, we use a measure of global coherence for the automated detection of linguistic items (e.g., sounds or morphemes) with higher or lesser language variation. The paper describes both the data and the method and provides analyses examples.