Mirage: 2D Source Localization Using Microphone Pair Augmentation with Echoes

Diego Di Carlo (1) , Antoine Deleforge (2) , Nancy Bertin (1)

1 PANAMA - Parcimonie et Nouveaux Algorithmes pour le Signal et la Modélisation Audio
2 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication

It is commonly observed that acoustic echoes hurt performance of sound source localization (SSL) methods. We introduce the concept of microphone array augmentation with echoes (MIRAGE) and show how estimation of early-echo characteristics can in fact benefit SSL.We propose a learningbased scheme for echo estimation combined with a physicsbased scheme for echo aggregation. In a simple scenario involving 2 microphones close to a reflective surface and one source, we show using simulated data that the proposed approach performs similarly to a correlation-based method in azimuth estimation while retrieving elevation as well from 2 microphones only, an impossible task in anechoic settings.

Mots clés: TDOA Estimation; Image Microphones; Sound Source Localization; Supervised Learning

Domaines: Applications [stat.AP]; Acoustique [physics.class-ph]; Traitement du signal et de l'image [eess.SP]

Diego Di Carlo, Antoine Deleforge, Nancy Bertin. Mirage: 2D Source Localization Using Microphone Pair Augmentation with Echoes. ICASSP 2019 - IEEE International Conference on Acoustic, Speech Signal Processing, May 2019, Brighton, United Kingdom. pp.775-779, ⟨10.1109/ICASSP.2019.8683534⟩. ⟨hal-02160940⟩