Branislav Gerazov

Daniel R. van Niekerk, Anqi Xu, Branislav Gerazov et al.

High-quality articulatory speech synthesis has many potential applications in speech science and technology. However, developing appropriate mappings from linguistic specification to articulatory gestures is difficult and time consuming. In this paper we construct an optimisation-based framework as a first step towards learning these mappings without manual intervention. We demonstrate the production of syllables with complex onsets and discuss the quality of the articulatory gestures with reference to coarticulation.

Stefan Krsteski, Matea Tashkovska, Borjan Sazdov et al.

The increase in technological adoption worldwide comes with demands for novel tools to be used by the general population. Large Language Models (LLMs) provide a great opportunity in this respect, but their capabilities remain limited for low-resource languages, restricting applications in countries where such languages are spoken. We create several resources to facilitate the adoption of LLMs and to support research advancements for Macedonian. We collect the largest Macedonian corpus to date, consisting of 40GB of textual data and totaling 3.5B words. To support conversational applications, we collect a 106k-instance instruction dataset, carefully built to be culturally grounded. For evaluation, we construct a Macedonian evaluation suite covering seven benchmarks. Finally, we train domestic-yak, a state-of-the-art 8B-parameter model, on our curated datasets and evaluate it against eight baseline models using the newly constructed benchmark suite. Our model outperforms all existing models in the 8B parameter range across all benchmarks, and achieves performance comparable to models up to 10x larger. Furthermore, a qualitative analysis with native speakers reveals that our model is preferred over larger counterparts, receiving higher ratings for grammatical correctness and cultural appropriateness. All datasets, code, and model weights are openly released, setting a foundation for advancing LLMs in similarly underrepresented languages. These resources are publicly available at github.com/LVSTCK for source code, and at huggingface.co/LVSTCK for pretrained model weights and data.

Branislav Gerazov, Marcello Politi, Sébastien Bratières

We explore the performance of several state-of-the-art automatic speech recognition (ASR) models on a large-scale Arabic speech dataset, the SADA (Saudi Audio Dataset for Arabic), which contains 668 hours of high-quality audio from Saudi television shows. The dataset includes multiple dialects and environments, specifically a noisy subset that makes it particularly challenging for ASR. We evaluate the performance of the models on the SADA test set, and we explore the impact of fine-tuning, language models, as well as noise and denoising on their performance. We find that the best performing model is the MMS 1B model finetuned on SADA with a 4-gram language model that achieves a WER of 40.9\% and a CER of 17.6\% on the SADA test clean set.

Mohammed Salah Al-Radhi, Géza Németh, Branislav Gerazov

Speech synthesis from intracranial EEG (iEEG) signals offers a promising avenue for restoring communication in individuals with severe speech impairments. However, achieving intelligible and natural speech remains challenging due to limitations in feature representation, prosody modeling, and phase reconstruction. We introduce MiSTR, a deep-learning framework that integrates: 1) Wavelet-based feature extraction to capture fine-grained temporal, spectral, and neurophysiological representations of iEEG signals, 2) A Transformer-based decoder for prosody-aware spectrogram prediction, and 3) A neural phase vocoder enforcing harmonic consistency via adaptive spectral correction. Evaluated on a public iEEG dataset, MiSTR achieves state-of-the-art speech intelligibility, with a mean Pearson correlation of 0.91 between reconstructed and original Mel spectrograms, improving over existing neural speech synthesis baselines.

Bojan Sofronievski, Branislav Gerazov

Music transcription is the process of transcribing music audio into music notation. It is a field in which the machines still cannot beat human performance. The main motivation for automatic music transcription is to make it possible for anyone playing a musical instrument, to be able to generate the music notes for a piece of music quickly and accurately. It does not matter if the person is a beginner and simply struggles to find the music score by searching, or an expert who heard a live jazz improvisation and would like to reproduce it without losing time doing manual transcription. We propose Scorpiano -- a system that can automatically generate a music score for simple monophonic piano melody tracks using digital signal processing. The system integrates multiple digital audio processing methods: notes onset detection, tempo estimation, beat detection, pitch detection and finally generation of the music score. The system has proven to give good results for simple piano melodies, comparable to commercially available neural network based systems.

Branislav Gerazov, Michael Wagner

The labelling of speech corpora is a laborious and time-consuming process. The ProsoBeast Annotation Tool seeks to ease and accelerate this process by providing an interactive 2D representation of the prosodic landscape of the data, in which contours are distributed based on their similarity. This interactive map allows the user to inspect and label the utterances. The tool integrates several state-of-the-art methods for dimensionality reduction and feature embedding, including variational autoencoders. The user can use these to find a good representation for their data. In addition, as most of these methods are stochastic, each can be used to generate an unlimited number of different prosodic maps. The web app then allows the user to seamlessly switch between these alternative representations in the annotation process. Experiments with a sample prosodically rich dataset have shown that the tool manages to find good representations of varied data and is helpful both for annotation and label correction. The tool is released as free software for use by the community.

Branislav Gerazov, Daniel van Niekerk, Anqi Xu et al.

The way infants use auditory cues to learn to speak despite the acoustic mismatch of their vocal apparatus is a hot topic of scientific debate. The simulation of early vocal learning using articulatory speech synthesis offers a way towards gaining a deeper understanding of this process. One of the crucial parameters in these simulations is the choice of features and a metric to evaluate the acoustic error between the synthesised sound and the reference target. We contribute with evaluating the performance of a set of 40 feature-metric combinations for the task of optimising the production of static vowels with a high-quality articulatory synthesiser. Towards this end we assess the usability of formant error and the projection of the feature-metric error surface in the normalised F1-F2 formant space. We show that this approach can be used to evaluate the impact of features and metrics and also to offer insight to perceptual results.

Branislav Gerazov, Gérard Bailly, Omar Mohammed et al.

The quest for comprehensive generative models of intonation that link linguistic and paralinguistic functions to prosodic forms has been a longstanding challenge of speech communication research. Traditional intonation models have given way to the overwhelming performance of deep learning (DL) techniques for training general purpose end-to-end mappings using millions of tunable parameters. The shift towards black box machine learning models has nonetheless posed the reverse problem -- a compelling need to discover knowledge, to explain, visualise and interpret. Our work bridges between a comprehensive generative model of intonation and state-of-the-art DL techniques. We build upon the modelling paradigm of the Superposition of Functional Contours (SFC) model and propose a Variational Prosody Model (VPM) that uses a network of variational contour generators to capture the context-sensitive variation of the constituent elementary prosodic contours. We show that the VPM can give insight into the intrinsic variability of these prosodic prototypes through learning a meaningful prosodic latent space representation structure. We also show that the VPM is able to capture prosodic phenomena that have multiple dimensions of context based variability. Since it is based on the principle of superposition, the VPM does not necessitate the use of specially crafted corpora for the analysis, opening up the possibilities of using big data for prosody analysis. In a speech synthesis scenario, the model can be used to generate a dynamic and natural prosody contour that is devoid of averaging effects.

Branislav Gerazov, Gérard Bailly, Yi Xu

The way speech prosody encodes linguistic, paralinguistic and non-linguistic information via multiparametric representations of the speech signals is still an open issue. The Superposition of Functional Contours (SFC) model proposes to decompose prosody into elementary multiparametric functional contours through the iterative training of neural network contour generators using analysis-by-synthesis. Each generator is responsible for computing multiparametric contours that encode one given linguistic, paralinguistic and non-linguistic information on a variable scope of rhythmic units. The contributions of all generators' outputs are then overlapped and added to produce the prosody of the utterance. We propose an extension of the contour generators that allows them to model the prominence of the elementary contours based on contextual information. WSFC jointly learns the patterns of the elementary multiparametric functional contours and their weights dependent on the contours' contexts. The experimental results show that the proposed weighted SFC (WSFC) model can successfully capture contour prominence and thus improve SFC modelling performance. The WSFC is also shown to be effective at modelling the impact of attitudes on the prominence of functional contours cuing syntactic relations in French, and that of emphasis on the prominence of tone contours in Chinese.