Fabio Tamburini

ORCID: 0000-0001-7950-0347
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Natural Language Processing Techniques
  • Linguistic Studies and Language Acquisition
  • Speech and dialogue systems
  • Topic Modeling
  • Phonetics and Phonology Research
  • Speech Recognition and Synthesis
  • Text Readability and Simplification
  • Semantic Web and Ontologies
  • Spanish Linguistics and Language Studies
  • Advanced Text Analysis Techniques
  • Speech and Audio Processing
  • Authorship Attribution and Profiling
  • Image Processing and 3D Reconstruction
  • Dementia and Cognitive Impairment Research
  • Quantum Computing Algorithms and Architecture
  • Linguistic Variation and Morphology
  • Real-Time Systems Scheduling
  • Open Education and E-Learning
  • Digital Communication and Language
  • Distributed and Parallel Computing Systems
  • Legal Language and Interpretation
  • Neural Networks and Applications
  • Linguistic Education and Pedagogy
  • Neurobiology of Language and Bilingualism
  • Eating Disorders and Behaviors

University of Bologna
2015-2025

Zambon (Italy)
2019

Società Italiana di Fisica
2019

ANT Foundation Italy Onlus
2019

Informa (Italy)
2016

University of Turin
2016

The discovery of early, non-invasive biomarkers for the identification "preclinical" or "pre-symptomatic" Alzheimer's disease and other dementias is a key issue, especially research purposes, design preventive clinical trials, drafting population-based health care policies. Complex behaviors are natural candidates this. In particular, recent studies have suggested that speech alterations might be one earliest signs cognitive decline, frequently noticeable years before deficits become...

10.3389/fnagi.2018.00369 article EN cc-by Frontiers in Aging Neuroscience 2018-11-13

10.1109/icassp49660.2025.10890866 article EN ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2025-03-12

Perceptual prominence is an important indicator of a word's and syllable's lexical, syntactic, semantic pragmatic status in discourse.Its automatic annotation would be valuable enrichment large databases used unit selection speech synthesis recognition.While much research has been carried out on the interaction between acoustic factors, little progress made its annotation.Previous approaches to German relied linguistic features detection, but purely method advantageous.We applied algorithm...

10.21437/interspeech.2007-505 article EN Interspeech 2022 2007-08-27

Prosodic prominence denotes the perceptual salience of linguistic units.There exists no agreement on (1) adequate methods for its subjective measurement, (2) objective acoustic correlates and (3) relationship to structure.A traditional approach evaluating any these descriptive layers is an inter-level comparison, e.g. between a model prominence.However, there standard procedure such comparison misleading if both are expected be symmetrical, given neglected influence top-down expectancies.We...

10.21437/interspeech.2012-625 article EN Interspeech 2022 2012-09-09

This paper presents a new approach for building Language Models using the Quantum Probability Theory, Model (QLM). It mainly shows that relying on this probability calculus it is possible to build stochastic models able benefit from quantum correlations due interference and entanglement. We extensively tested our showing its superior performances, both in terms of model perplexity inserting into an automatic speech recognition evaluation setting, when compared with state-of-the-art language...

10.18653/v1/d17-1196 article EN cc-by Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing 2017-01-01

This paper presents a follow up of study on the automatic detection prosodic prominence in spontaneous speech.Prosodic involves two different features, pitch accent and stress, that are typically based four acoustic parameters: fundamental frequency (F0) movements, overall syllable energy, nuclei duration mid-tohigh-frequency emphasis.A careful measurement these parameters makes it possible to build an system capable identifying prominent syllables utterances with performance comparable...

10.21437/speechprosody.2006-18 article EN Speech prosody 2006-05-02

This paper presents work in progress on the automatic detection of prosodic prominence continuous speech. Prosodic involves two different phonetic features: pitch accents, connected with fundamental frequency (FO) movements and syllable overall energy, stress, which exhibits a strong correlation nuclei duration high-frequency emphasis. By measuring these acoustic parameters it is possible to build an system capable correctly identifying prominent syllables agreement human-tagged data...

10.1109/isspa.2003.1224721 article EN 2003-01-01

This paper presents work in progress on the automatic detection of prosodic prominence continuous speech. Prosodic involves two different phonetic features: pitch accents, connected with fundamental frequency (F0) movements and syllable overall energy, stress, which exhibits a strong correlation nuclei duration mid-to-high-frequency emphasis. By measuring these acoustic parameters it is possible to build an system capable correctly identifying prominent syllables agreement, human-tagged...

10.21437/eurospeech.2003-81 article EN 2003-09-01

Prosodic prominence, a speech phenomenon by which some linguistic units are perceived as standing out from their environment, plays very important role in human communication. In this paper we present study on automatic prominence identification using Probabilistic Graphical Models, family of Machine Learning Systems able to properly handle sequences events. We tested the most promising members such models utterances selected manually annotated Italian corpus, obtaining good recognition...

10.21437/speechprosody.2014-45 article EN Speech prosody 2014-05-20

Ancient undeciphered scripts present problems of different nature, not just tied to linguistic identification. The Cypro-Minoan script from second millennium BCE Cyprus, for instance, currently does have a standardized, definitive inventory signs, and, in addition, stands divided into three separate subgroups (CM1, CM2, CM3), which also been alleged record languages. However, this state the art is consensually accepted by experts. In article, we aim apply method that can aid shed light on...

10.1371/journal.pone.0269544 article EN cc-by PLoS ONE 2022-07-14

This paper presents a follow up of study on the automatic detection prosodic prominence in continuous speech. Prosodic involves two different features, pitch accent and stress, that are typically based four acoustic parameters: fundamental frequency (F0) movements, overall syllable energy, nuclei duration mid-tohigh-frequency emphasis. A careful measurement these parameters, as well identification their connection to makes it possible build an system capable identifying prominent syllables...

10.21437/interspeech.2005-553 article EN Interspeech 2022 2005-09-04

Minoan Linear A is still an undeciphered script mainly used for administrative purposes on Bronze Age Crete. One of its most enigmatic features the precise mathematical values system numerical fractions. The aim this article to address issue through a multi-stranded methodology that comprises palaeographical examination and statistical, computational, typological approaches. Taking from previous analyses, which suggested hypothetical some fractions, we extended our probe into assessing...

10.1016/j.jas.2020.105214 article EN cc-by-nc-nd Journal of Archaeological Science 2020-09-07
Coming Soon ...