arXiv subjects

Andreas Triantafyllopoulos

Publications and source records attributed to Andreas Triantafyllopoulos.

At least 19 recordsLinked to original sources

arXiv2026

Is Natural Always Appropriate? Investigating Naturalness and Appropriateness Across Different Domains for TTS Evaluation

Text-to-speech (TTS) evaluation is an open challenge. While the primary target was "naturalness," recent fidelity gains shifted focus toward "appropriateness" and whether speech is correct for its context. In this work, we examine how perception changes when the expected downstream use varies. We measure the appropriateness and human-likeness of five SOTA TTS systems across five domains: AI assistant, reader, actor, animated character, and spontaneous speaker. Results show appropriateness varies across domains independently of naturalness. While systems shine at reading, expressive domains remain challenging, and optimizing for one can degrade others. Furthermore, naturalness scores tend to penalize stylized speech while rewarding spontaneity. Finally, our study also highlights blind spots in one-size-fits-all evaluation metrics across more expressive domains. We demonstrate that TTS performance is not "solved" but depends on the target domain, requiring context-aware evaluation.

eess.AS↗

arXiv2026

CoughPhase-CLR: Designing an acoustics-informed foundation model for coughing sound classification

In this work, we introduce CoughPhase-CLR, a self-supervised learning framework designed to leverage the physiological phases of a cough for robust representation learning. Unlike generic contrastive frameworks, CoughPhase-CLR constructs positive pairs based on these specific acoustic phases. We pre-trained our model on approximately 40 hours of public cough audio and evaluated it across five downstream tasks, including COVID-19 detection, chronic obstructive pulmonary disease (COPD) state classification, and smoker status prediction. Our results demonstrate that cough-specific pre-training consistently outperforms standard random-cropping techniques when training on cough recordings. Additionally, we benchmarked a diverse set of state-of-the-art models on COPD state classification, highlighting the difficulty of this task. The best-performing models, pretrained on either general audio or respiratory sounds, achieved a UAR of 57\%, failing to outperform the state-of-the-art performance of 84\% UAR achieved using speech analysis.

cs.SD↗

arXiv2026

Acoustic Cue Alignment in Audio Language Models for Speech Emotion Recognition

Instruction-following audio language models (ALMs) can be augmented with explicit acoustic cues, yet it remains unclear whether such cues are used in a grounded way when the raw audio is already available. We study this question in speech emotion recognition (SER) by deriving six interpretable acoustic concept tokens from the standardised eGeMAPS paralinguistic feature set. These tokens summarise energy, pitch, dynamics, brightness, formants, and voice quality, and are appended to the textual prompt while the audio input is kept unchanged. Across the widely used FAU-Aibo and IEMOCAP benchmarks, aligned tokens improve unweighted average recall (UAR), whereas shuffled, conflicting, or corrupted tokens reduce performance relative to aligned tokens and shift confusions toward neutral. Importantly, predictions do not collapse under strong token perturbations, suggesting that the models are sensitive to the symbolic cue channel but remain partly anchored to the audio signal. We argue that token-only interventions provide a practical way to probe audio-grounded cue use, robustness, and interpretability in ALM-based affective computing.

cs.SD↗

arXiv2026

A Pilot Study on Curator-Guided Multilingual Art Description for Blind and Low-Vision Audiences with Small Vision-Language Models

Blind and low-vision (BLV) audiences remain underserved by visual art descriptions, particularly across languages and in museum settings where privacy and intellectual-property constraints may favour small on-premise vision-language models (VLMs). This pilot study investigates curator-guided multilingual art description with Qwen2.5-VL-3B-Instruct for German, Romanian, and Serbian. We construct a parallel BLV-oriented caption corpus from artwork images and metadata, and compare language-specific LoRA adapters with a single multilingual adapter under a fixed backbone and training budget. Evaluation combines automatic lexical and embedding-based metrics with an LLM-as-Judge protocol calibrated against a small Romanian BLV pilot study. Under our pilot setup, language-specific adapters show more stable controllability and visually grounded description quality for Romanian and Serbian, while multilingual adaptation remains competitive in German. We frame these findings as deployment-oriented evidence for small on-premise VLMs, and highlight the need for larger BLV user studies and broader language coverage before drawing general conclusions about multilingual accessibility.

cs.MM↗

arXiv2026

CoarseSoundNet: Building a reliable model for ecological soundscape analysis

A soundscape is composed of three types of sound: biophony (sounds made by animals), geophony (natural abiotic sounds) and anthropophony (sounds made by humans). A key research question in the field of soundscape ecology is how these components interact with each other, specifically how biophony responds to geophony and anthropophony. Nevertheless, as of today, there are not many analytical instruments that enable the distinct quantification of these elements. Recent machine learning (ML) approaches aim to support automated analysis but often rely on task-specific or clean data, limiting generalisation to noisy passive acoustic monitoring (PAM) recordings. This study presents a clear and reproducible structure to build ML models for coarse soundscape classification and introduces CoarseSoundNet, a deep learning model trained to distinguish biophony, geophony, and anthropophony under realistic PAM conditions. We systematically investigate model architectures, the influence of an additional training class, data composition, and evaluation strategies. Our findings suggest that model performance improves with additional PAM data, especially when similar to the target domain, and by introducing an explicit silence class during training. Class-specific decision thresholds and duration-based constraints further enhance performance, particularly for anthropophony and geophony. Error analyses exhibit challenges for anthropophony due to masking effects and confusions for silence and insect sounds for geophony and biophony. Finally, we conduct an ecological case study which shows that pre-filtering recordings with CoarseSoundNet yields acoustic index trends comparable to ground-truth filtering, supporting its use as an effective preprocessing tool for ecoacoustic analyses.

cs.SD↗

arXiv2026

A conceptual framework for learning to listen by reward: Curiosity-driven search for novel sources

Reinforcement learning is a powerful learning paradigm that has spearheaded progress in numerous domains. Its core promise lies in learning through high-level goals without the need for granular labels. However, it still remains elusive in the realm of audio, where it has received substantially less attention than in computer vision or other domains. The key question remains: how can agents learn to listen purely via reward-driven exploration? In this contribution, we present an overview of previous attempts and a new conceptual framework for learning to listen by reward. Our approach depends on the continuous search for novel sound sources. We formulate our framework, discuss open technical challenges, and present a first proof-of-concept implementation that showcases the feasibility of our approach.

cs.SD↗

arXiv2026

How Class Ontology and Data Scale Affect Audio Transfer Learning

Transfer learning is a crucial concept within deep learning that allows artificial neural networks to benefit from a large pre-training data basis when confronted with a task of limited data. Despite its ubiquitous use and clear benefits, there are still many open questions regarding the inner workings of transfer learning and, in particular, regarding the understanding of when and how well it works. To that extent, we perform a rigorous study focusing on audio-to-audio transfer learning, in which we pre-train various model states on (ontology-based) subsets of AudioSet and fine-tune them on three computer audition tasks, namely acoustic scene recognition, bird activity recognition, and speech command recognition. We report that increasing the number of samples and classes in the pre-training data both have a positive impact on transfer learning. This is, however, generally surpassed by similarity between pre-training and the downstream task, which can lead the model to learn comparable features.

cs.LG↗

arXiv2026

EmoSURA: Towards Accurate Evaluation of Detailed and Long-Context Emotional Speech Captions

Recent advancements in speech captioning models have enabled the generation of rich, fine-grained captions for emotional speech. However, the evaluation of such captions remains a critical bottleneck: traditional N-gram metrics fail to capture semantic nuances, while LLM judges often suffer from reasoning inconsistency and context-collapse when processing long-form descriptions. In this work, we propose EmoSURA, a novel evaluation framework that shifts the paradigm from holistic scoring to atomic verification. EmoSURA decomposes complex captions into Atomic Perceptual Units, which are self-contained statements regarding vocal or emotional attributes, and employs an audio-grounded verification mechanism to validate each unit against the raw speech signal. Furthermore, we address the scarcity of standardized evaluation resources by introducing SURABench, a carefully balanced and stratified benchmark. Our experiments show that EmoSURA achieves a positive correlation with human judgments, offering a more reliable assessment for long-form captions compared to traditional metrics, which demonstrated negative correlations due to their sensitivity to caption length.

cs.SD↗

arXiv2026

SmoothCLAP: Soft-Target Enhanced Contrastive Language\--Audio Pretraining for Affective Computing

The ambiguity of human emotions poses several challenges for machine learning models, as they often overlap and lack clear delineating boundaries. Contrastive language-audio pretraining (CLAP) has emerged as a key technique for generalisable emotion recognition. However, as conventional CLAP enforces a strict one-to-one alignment between paired audio-text samples, it overlooks intra-modal similarity and treats all non-matching pairs as equally negative. This conflicts with the fuzzy boundaries between different emotions. To address this limitation, we propose SmoothCLAP, which introduces softened targets derived from intra-modal similarity and paralinguistic features. By combining these softened targets with conventional contrastive supervision, SmoothCLAP learns embeddings that respect graded emotional relationships, while retaining the same inference pipeline as CLAP. Experiments on eight affective computing tasks across English and German demonstrate that SmoothCLAP is consistently achieving superior performance. Our results highlight that leveraging soft supervision is a promising strategy for building emotion-aware audio-text models.

cs.SD↗

arXiv2025

Detecting COPD Through Speech Analysis: A Dataset of Danish Speech and Machine Learning Approach

Chronic Obstructive Pulmonary Disease (COPD) is a serious and debilitating disease affecting millions around the world. Its early detection using non-invasive means could enable preventive interventions that improve quality of life and patient outcomes, with speech recently shown to be a valuable biomarker. Yet, its validity across different linguistic groups remains to be seen. To that end, audio data were collected from 96 Danish participants conducting three speech tasks (reading, coughing, sustained vowels). Half of the participants were diagnosed with different levels of COPD and the other half formed a healthy control group. Subsequently, we investigated different baseline models using openSMILE features and learnt x-vector embeddings. We obtained a best accuracy of 67% using openSMILE features and logistic regression. Our findings support the potential of speech-based analysis as a non-invasive, remote, and scalable screening tool as part of future COPD healthcare solutions.

cs.SD↗

arXiv2025

Charting 15 years of progress in deep learning for speech emotion recognition: A replication study

Speech emotion recognition (SER) has long benefited from the adoption of deep learning methodologies. Deeper models -- with more layers and more trainable parameters -- are generally perceived as being `better' by the SER community. This raises the question -- \emph{how much better} are modern-era deep neural networks compared to their earlier iterations? Beyond that, the more important question of how to move forward remains as poignant as ever. SER is far from a solved problem; therefore, identifying the most prominent avenues of future research is of paramount importance. In the present contribution, we attempt a quantification of progress in the 15 years of research beginning with the introduction of the landmark 2009 INTERSPEECH Emotion Challenge. We conduct a large scale investigation of model architectures, spanning both audio-based models that rely on speech inputs and text-baed models that rely solely on transcriptions. Our results point towards diminishing returns and a plateau after the recent introduction of transformer architectures. Moreover, we demonstrate how perceptions of progress are conditioned on the particular selection of models that are compared. Our findings have important repercussions about the state-of-the-art in SER research and the paths forward

cs.SD↗

arXiv2025

Reading Smiles: Proxy Bias in Foundation Models for Facial Emotion Recognition

Foundation Models (FMs) are rapidly transforming Affective Computing (AC), with Vision Language Models (VLMs) now capable of recognising emotions in zero shot settings. This paper probes a critical but underexplored question: what visual cues do these models rely on to infer affect, and are these cues psychologically grounded or superficially learnt? We benchmark varying scale VLMs on a teeth annotated subset of AffectNet dataset and find consistent performance shifts depending on the presence of visible teeth. Through structured introspection of, the best-performing model, i.e., GPT-4o, we show that facial attributes like eyebrow position drive much of its affective reasoning, revealing a high degree of internal consistency in its valence-arousal predictions. These patterns highlight the emergent nature of FMs behaviour, but also reveal risks: shortcut learning, bias, and fairness issues especially in sensitive domains like mental health and education.

cs.CV↗

arXiv2025

MELT: Towards Automated Multimodal Emotion Data Annotation by Leveraging LLM Embedded Knowledge

Although speech emotion recognition (SER) has advanced significantly with deep learning, annotation remains a major hurdle. Human annotation is not only costly but also subject to inconsistencies annotators often have different preferences and may lack the necessary contextual knowledge, which can lead to varied and inaccurate labels. Meanwhile, Large Language Models (LLMs) have emerged as a scalable alternative for annotating text data. However, the potential of LLMs to perform emotional speech data annotation without human supervision has yet to be thoroughly investigated. To address these problems, we apply GPT-4o to annotate a multimodal dataset collected from the sitcom Friends, using only textual cues as inputs. By crafting structured text prompts, our methodology capitalizes on the knowledge GPT-4o has accumulated during its training, showcasing that it can generate accurate and contextually relevant annotations without direct access to multimodal inputs. Therefore, we propose MELT, a multimodal emotion dataset fully annotated by GPT-4o. We demonstrate the effectiveness of MELT by fine-tuning four self-supervised learning (SSL) backbones and assessing speech emotion recognition performance across emotion datasets. Additionally, our subjective experiments\' results demonstrate a consistence performance improvement on SER.

cs.AI↗

arXiv2025

ECOSoundSet: a finely annotated dataset for the automated acoustic identification of Orthoptera and Cicadidae in North, Central and temperate Western Europe

Currently available tools for the automated acoustic recognition of European insects in natural soundscapes are limited in scope. Large and ecologically heterogeneous acoustic datasets are currently needed for these algorithms to cross-contextually recognize the subtle and complex acoustic signatures produced by each species, thus making the availability of such datasets a key requisite for their development. Here we present ECOSoundSet (European Cicadidae and Orthoptera Sound dataSet), a dataset containing 10,653 recordings of 200 orthopteran and 24 cicada species (217 and 26 respective taxa when including subspecies) present in North, Central, and temperate Western Europe (Andorra, Belgium, Denmark, mainland France and Corsica, Germany, Ireland, Luxembourg, Monaco, Netherlands, United Kingdom, Switzerland), collected partly through targeted fieldwork in South France and Catalonia and partly through contributions from various European entomologists. The dataset is composed of a combination of coarsely labeled recordings, for which we can only infer the presence, at some point, of their target species (weak labeling), and finely annotated recordings, for which we know the specific time and frequency range of each insect sound present in the recording (strong labeling). We also provide a train/validation/test split of the strongly labeled recordings, with respective approximate proportions of 0.8, 0.1 and 0.1, in order to facilitate their incorporation in the training and evaluation of deep learning algorithms. This dataset could serve as a meaningful complement to recordings already available online for the training of deep learning algorithms for the acoustic classification of orthopterans and cicadas in North, Central, and temperate Western Europe.

cs.SD↗

arXiv2025

DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Aids

The DeepFilterNet (DFN) architecture was recently proposed as a deep learning model suited for hearing aid devices. Despite its competitive performance on numerous benchmarks, it still follows a `one-size-fits-all' approach, which aims to train a single, monolithic architecture that generalises across different noises and environments. However, its limited size and computation budget can hamper its generalisability. Recent work has shown that in-context adaptation can improve performance by conditioning the denoising process on additional information extracted from background recordings to mitigate this. These recordings can be offloaded outside the hearing aid, thus improving performance while adding minimal computational overhead. We introduce these principles to the DFN model, thus proposing the DFingerNet (DFiN) model, which shows superior performance on various benchmarks inspired by the DNS Challenge.

cs.SD↗

arXiv2024

autrainer: A Modular and Extensible Deep Learning Toolkit for Computer Audition Tasks

This work introduces the key operating principles for autrainer, our new deep learning training framework for computer audition tasks. autrainer is a PyTorch-based toolkit that allows for rapid, reproducible, and easily extensible training on a variety of different computer audition tasks. Concretely, autrainer offers low-code training and supports a wide range of neural networks as well as preprocessing routines. In this work, we present an overview of its inner workings and key capabilities.

cs.SD↗

arXiv2024

Large language models for mental health

Digital technologies have long been explored as a complement to standard procedure in mental health research and practice, ranging from the management of electronic health records to app-based interventions. The recent emergence of large language models (LLMs), both proprietary and open-source ones, represents a major new opportunity on that front. Yet there is still a divide between the community developing LLMs and the one which may benefit from them, thus hindering the beneficial translation of the technology into clinical use. This divide largely stems from the lack of a common language and understanding regarding the technology's inner workings, capabilities, and risks. Our narrative review attempts to bridge this gap by providing intuitive explanations behind the basic concepts related to contemporary LLMs.

cs.CY↗

arXiv2024

Does the Definition of Difficulty Matter? Scoring Functions and their Role for Curriculum Learning

Curriculum learning (CL) describes a machine learning training strategy in which samples are gradually introduced into the training process based on their difficulty. Despite a partially contradictory body of evidence in the literature, CL finds popularity in deep learning research due to its promise of leveraging human-inspired curricula to achieve higher model performance. Yet, the subjectivity and biases that follow any necessary definition of difficulty, especially for those found in orderings derived from models or training statistics, have rarely been investigated. To shed more light on the underlying unanswered questions, we conduct an extensive study on the robustness and similarity of the most common scoring functions for sample difficulty estimation, as well as their potential benefits in CL, using the popular benchmark dataset CIFAR-10 and the acoustic scene classification task from the DCASE2020 challenge as representatives of computer vision and computer audition, respectively. We report a strong dependence of scoring functions on the training setting, including randomness, which can partly be mitigated through ensemble scoring. While we do not find a general advantage of CL over uniform sampling, we observe that the ordering in which data is presented for CL-based training plays an important role in model performance. Furthermore, we find that the robustness of scoring functions across random seeds positively correlates with CL performance. Finally, we uncover that models trained with different CL strategies complement each other by boosting predictive power through late fusion, likely due to differences in the learnt concepts. Alongside our findings, we release the aucurriculum toolkit (https://github.com/autrainer/aucurriculum), implementing sample difficulty and CL-based training in a modular fashion.

cs.LG↗