Blind Acoustic Parameter Estimation Through Task-Agnostic Embeddings Using Latent Approximations

2024/07/29 by Philipp Götz, Götz, Philipp, Çağdaş Tuna +7 · 2 citations
Computer Science · #Audio and Speech Processing (eess.AS) #FOS: Electrical engineering #Music and Audio Processing #Speech Recognition and Synthesis #Speech and Audio Processing #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2407.19989

openalex publication_date 2024/07/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present a method for blind acoustic parameter estimation from single-channel reverberant speech. The method is structured into three stages. In the first stage, a variational auto-encoder is trained to extract latent representations of acoustic impulse responses represented as mel-spectrograms. In the second stage, a separate speech encoder is trained to estimate low-dimensional representations from short segments of reverberant speech. Finally, the pre-trained speech encoder is combined with a small regression model and evaluated on two parameter regression tasks. Experimentally, the proposed method is shown to outperform a fully end-to-end trained baseline model.

Cited by

• Matching Reverberant Speech Through Learned Acoustic Embeddings and Feedback Delay Networks
• Loss functions incorporating auditory spatial perception in deep learning -- a review

Related

• Contrastive Representation Learning for Acoustic Parameter Estimation
• Impulse Response Data Augmentation and Deep Neural Networks for Blind Room Acoustic Parameter Estimation
• Realistic sources, receivers and walls improve the generalisability of virtually-supervised blind acoustic parameter estimators
• Joint Estimation of Reverberation Time and Direct-to-Reverberation Ratio from Speech using Auditory-Inspired Features
• Joint Blind Room Acoustic Characterization From Speech And Music Signals Using Convolutional Recurrent Neural Networks