cs.SDNov 9, 2026

Reverberant Speech Conditioning for Material-Aware Acoustic Parameter Mapping

Authors: Yi Wu, Olha Townsend, Qianqian Fang, Magdalena Fuentes, You Zhang

Organizations: Dolby Laboratories, Inc. · New York University

Abstract

Dense maps of room acoustic parameters (T30, EDT, C50, DRR) describe how a room sounds at every position. Scene-wide estimation predicts such maps from a floor plan but requires a reference impulse response, which is hard to obtain. Our system instead receives a floor plan and easily recorded reverberant speech from one talker-microphone pair. We compare four speech-derived conditioning inputs: MFCCs, an embedding from the blind estimator BERP, BERP's scalar reverberation-time estimate, and an impulse response reconstructed by Rec-RIR. Our main benchmark is the MRAS dataset, where material assignments vary at fixed geometry. The reconstructed impulse response performs best among the four inputs, reducing T30 error from 31.5% with the floor plan alone to 19.9%, 83% of the reduction achieved by the reference response. We find that adding speech mainly sets the overall level of the map, with little improvement in predicting how the acoustics vary across the room.

Figures & tables

Explore similar work

Oct 30, 2026

ARPEGE: Dataset and Methods to Estimate Acoustic and Geometric Parameters of Rooms from Sound

Can one hear the walls of a room? To foster audio signal processing research around this question we release ARPEGE, a dataset containing hundreds of ambisonic spherical-array room impulse responses recorded in 6 empty rooms, annotated by the relative positions of sources, microphones and all acoustic reflectors. The dataset includes multiple room sizes, surface materials, orientations and distances, as well as recordings of human speakers and handclaps. Two state-of-the-art methods on image source localization and blind room parameter estimation are demonstrated on the data. Initial code for re-simulating the measured room impulse responses is provided, a longer term goal being to digitally twin the dataset.
Sep 25, 2026

Room Impulse Response Embeddings for Speech Enhancement in Noisy and Reverberant Environments

We propose a self-supervised approach for learning room impulse response (RIR) representations from single-channel noisy-reverberant speech. It consists of first training on reverberant data, then on noisy-reverberant data, and finally with a teacher-student approach, where the student learns to replicate the teacher's embeddings when given a noisy version of the reverberant input. We assess their representational capabilities by estimating acoustic room parameters from them. Conditioning a discriminative speech enhancement model on the derived embeddings yields consistent gains across all evaluated metrics, including downstream word error rate, for both reverberant and noisy-reverberant speech.
Feb 27, 2026

Whisper-RIR-Mega: A Paired Clean-Reverberant Speech Benchmark for ASR Robustness to Room Acoustics

We introduce Whisper-RIR-Mega, a benchmark dataset of paired clean and reverberant speech for evaluating automatic speech recognition (ASR) robustness to room acoustics. Each sample pairs a clean LibriSpeech utterance with the same utterance convolved with a real room impulse response from the RIR-Mega corpus, with stratified splits by reverberation time (RT60) and direct-to-reverberant ratio (DRR). We evaluate five Whisper models (tiny through large-v3) on 1600 test samples and report word error rate (WER) and character error rate (CER) under clean and reverberant conditions. Reverberation consistently degrades performance across all model sizes; the reverb penalty in WER ranges from 2.31 to 15.50 percentage points depending on the model. Whisper-large-v3 shows the smallest penalty; Whisper-tiny shows the largest. We release the dataset, evaluation code, and baseline results to support reproducible research on robust ASR.