A5076431415- Music and Audio Processing
- Music Technology and Sound Studies
- Speech and Audio Processing
- Ichthyology and Marine Biology
- Model Reduction and Neural Networks
- Diverse Musicological Studies
- Video Analysis and Summarization
- Identification and Quantification in Food
- Morphological variations and asymmetry
- Hearing Loss and Rehabilitation
- Fish Biology and Ecology Studies
- Artificial Intelligence in Games
- Educational Games and Gamification
- Fish Ecology and Management Studies
- Developmental Biology and Gene Regulation
- Teaching and Learning Programming
- Neural Networks and Applications
- Digital Filter Design and Implementation
- Advanced Adaptive Filtering Techniques
- Analog and Mixed-Signal Circuit Design
- Neuroscience and Music Perception
- Image Processing and 3D Reconstruction
- Multisensory perception and integration
- Speech Recognition and Synthesis
- Paleontology and Evolutionary Biology
Université de Bordeaux
2016-2022
Laboratoire Bordelais de Recherche en Informatique
2016-2022
Centre National de la Recherche Scientifique
2016-2022
Institut Polytechnique de Bordeaux
2017-2022
Institut de Neurosciences Cognitives et Intégratives d’Aquitaine
2017
Scyland3D is a Python tool for converting 3D raw landmark and semilandmark coordinates exported from acquisition software to CSV best suited geometric morphometric analyses.Processing schemes mirror reorder these points are provided address further symmetry issues.This useful when working on left-right differences in given species.This designed pre-process the extracted landmarks generate database-like statistical analyses.The main targeted audience biologists looking an easy quick parser landmarks.
Shark populations that are distributed alongside a latitudinal gradient often display body size differences at sexual maturity and vicariance patterns related to their number of tooth files. Previous works have demonstrated Scyliorhinus canicula differ between the northeastern Atlantic Ocean Mediterranean Sea based on biological features genetic analysis. In this study, we sample more than 3,000 teeth from 56 S. specimens caught incidentally off Roscoff Banyuls-sur-Mer. We investigate...
We introduce Kara1k, a new musical dataset composed of 2,000 analyzed songs thanks to partnership with karaoke company. The is divided into 1,000 cover provided by Recisio Karafun application1, and the corresponding original artists. Kara1k mainly dedicated toward song identification singing voice analysis. For both tasks, it offers novel approaches, as each studio-recorded same arrangement recording, but different singers musicians. Essentia, harmony-analyser, Marsyas, Vamp plugins YAAFE...
Abstract Regionalization of the vertebral column occurred early during vertebrate evolution and has been extensively investigated in mammals. However, less data are available on regions crown gnathostomes. This is particularly true for batoids (skates, sawfishes, guitarfishes, rays) whose long considered to be composed same two as teleost fishes despite presence a synarcual. numerous units chondrichthyans may display more complex regionalization pattern than previously assumed intraspecific...
This paper introduces SATIN, the Set of Audio Tags and Identifiers Normalized. SATIN is a database 400k audio-related metadata identifiers that aims at facilitating reproducibility comparisons among MIR algorithms. The idea to take advantage partnerships between scientists private companies host millions tracks. Scientists can send their feature extraction algorithm along SATIN's retrieve corresponding features. procedure allows community have access more tracks for classification purposes....
We introduce KaraMIR, a musical project dedicated to karaoke song analysis. Within we define Kara1k, dataset composed of 1000 cover songs provided by Recisio Karafun application, and the corresponding original artists. Kara1k is mainly toward identification singing voice For both tasks, offers novel approaches, as each studio-recorded with same arrangement recording, but different singers musicians. Essentia, harmony-analyser, Marsyas, Vamp plugins YAAFE have been used extract audio features...
This study deals with content-based musical playlists generation focused on Songs and Instrumentals. Automatic playlist relies collaborative filtering autotagging algorithms. Autotagging can solve the cold start issue popularity bias that are critical in music recommender systems. However, remains to be improved cannot generate satisfying playlists. In this paper, we suggest improvements toward better autotagging-generated compared state-of-the-art. To assess our method, focus Song...