Author: Vincent Lostanlen

Hold Me Tight: Stable Encoder–Decoder Design for Speech Enhancement

Convolutional layers with 1-D filters are often used as frontend to encode audio signals. Unlike fixed time-frequency representations, they can adapt to the local characteristics of input data. However, 1-D filters on raw audio are hard to train and often suffer from instabilities. In this paper, we address these problems with hybrid solutions, i.e., combining theory-driven and datadriven approaches. First, we preprocess the audio signals via a auditory filterbank, guaranteeing good frequency localization for the learned encoder. Second, we use results from frame theory to define an unsupervised learning objective that encourages energy conservation and perfect reconstruction. Third, we adapt mixed compressed spectral norms as learning objectives to the encoder coefficients. Using these solutions in a low-complexity encoder-mask-decoder model significantly improves the perceptual evaluation of speech quality (PESQ) in speech enhancement.

Machine Listening in a Neonatal Intensive Care Unit @ DCASE

Oxygenators, alarm devices, and footsteps are some of the most common sound sources in a hospital. Detecting them has scientific value for environmental psychology but comes with challenges of its own: namely, privacy preservation and limited labeled data. In this paper, we address these two challenges via a combination of edge computing and cloud computing. For privacy preservation, we have designed an acoustic sensor which computes third-octave spectrograms on the fly instead of recording audio waveforms. For sample-efficient machine learning, we have repurposed a pretrained audio neural network (PANN) via spectral transcoding and label space adaptation. A small-scale study in a neonatological intensive care unit (NICU) confirms that the time series of detected events align with another modality of measurement: i.e., electronic badges for parents and healthcare professionals. Hence, this paper demonstrates the feasibility of polyphonic machine listening in a hospital ward while guaranteeing privacy by design.

Journée GdR IASIS “Synthèse audio” à l’Ircam

As part of the CNRS special interest group on signal and image processing (GdR IASIS), we are organizing a 1-day workshop on audio synthesis at Ircam on November 7th, 2024.

Dans le cadre de l’axe « Audio, Vision, Perception » du GdR IASIS, nous organisons une journée d’études dédiée à la synthèse audio. La journée se tiendra le jeudi 7 novembre 2024 à l’Ircam (laboratoire STMS), à Paris.

Trainable signal encoders that are robust against noise @ Inter-Noise

Within the deep learning paradigm, finite impulse response (FIR) filters are often used to encode audio signals, yielding flexible and adaptive feature representations. We show that a stabilization of FIR filterbanks with fixed filter lengths (convolutional layers with 1-D filters)leads to encoders that are optimally robust against noise and can be inverted with perfect reconstruction by their transposes. To maintain their flexibility as regular neural network layers, we implement the stabilization via a computationally efficient regularizing term in the objective function of the learning problem. In this way, the encoder keeps its expressive power and is optimally stable and noise-robust throughout the whole learning procedure. We show in a denoising task where noise is present in the input and in the encoder representation, that the proposed stabilization of the trainable filterbank encoder is decisive for increasing the signal-to-noise ratio of the denoised signals significantly compared to a model with a naively trained encoder.

Journées du réseau “Capteurs en environnement”

Le comité d’organisation de ces journées est heureux de vous accueillir du 13 au 15 novembre 2024 à Fréjus, dans le Var, au centre de vacances de la Villa Clythia.

Ces journées, réservées en priorité aux membres du réseau, seront l’occasion de découvrir le travail des membres du réseau, de se rencontrer et d’échanger sur les nombreuses thématiques abordées par le réseau, depuis le développement de capteurs jusqu’à l’acquisition et le traitement des données, en passant par les contraintes de déploiement in natura.

Table ronde “IA en recherche” @ Ifremer

Cet événement, organisé par les doctorants de l’Ifremer, a déjà rassemblé une centaine de participants chaque année autour de présentations orales, de tables rondes et d’ateliers de prise en main. Cette journée est ouverte à toute personne intéressée par l’IA dans la recherche, des applications, les limites et les aspects éthiques.
Rendez-vous le mardi 8 octobre 2024 dans le centre Atlantique de l’Ifremer à Nantes.

BirdVoxDetect: Large-Scale Detection and Classification of Flight Calls for Bird Migration Monitoring @ IEEE TASLP

Sound event classification has the potential to advance our understanding of bird migration. Although it is long known that migratory species have a vocal signature of their own, previous work on automatic flight call classification has been limited in robustness and scope: e.g., covering few recording sites, short acquisition segments, and simplified biological taxonomies. In this paper, we present BirdVoxDetect (BVD), the first full-fledged solution to bird migration monitoring from acoustic sensor network data.

Loïc demonstrates his FTM Synthesizer on Erae Touch

A demonstration by independent researcher Loïc Jankowiak, who visited LS2N in 2023 and 2024 as part of the ApEx project of Han Han.

This is an in-progress build of a physically-modelled drum synthesizer using the Functional Transformation Method (FTM), with external MIDI control. The Embodme Erae Touch control surface is used for the velocity and impact position on the virtual drum’s surface, and the Behringer X-Touch Mini is used for controlling the various parameters of the physical model, i.e. the material properties.