DOA estimation of audio sources in reverberant environments

Jensen, Jesper Rindom; Nielsen, Jesper Kjær; Heusdens, Richard; Christensen, Mads Græsbøll

doi:10.1109/icassp.2016.7471660

Cited by 15 publications

(7 citation statements)

References 23 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practice however, modeling and estimation errors typically occur in all quantities. First, the RTF vector may differ from the true RTF vector, e.g., due to DOA estimation errors in highly reverberant and noisy scenarios [39]- [42]. Second, since the spatial coherence matrix is typically computed assuming a perfectly diffuse sound field for the late reverberation whereas this is not the case in practice, it typically differs from the true spatial coherence matrix.…”

Section: Impact Of Modeling and Estimation Errors On The Ml-basedmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of Eigenvalue Decomposition-Based Late Reverberation Power Spectral Density Estimation

Kodrasi¹,

Doclo²

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Many speech dereverberation techniques require an estimate of the late reverberation power spectral density (PSD). State-of-the-art multi-channel methods for estimating the late reverberation PSD typically rely on 1) an estimate of the relative transfer functions (RTFs) of the target signal, 2) a model for the spatial coherence matrix of the late reverberation, and 3) an estimate of the reverberant speech or reverberant and noisy speech PSD matrix. The RTFs, the spatial coherence matrix, and the speech PSD matrix are all prone to modeling and estimation errors in practice, with the RTFs being particularly difficult to estimate accurately, especially in highly reverberant and noisy scenarios. Recently, we proposed an eigenvalue decomposition (EVD)-based late reverberation PSD estimator which does not require an estimate of the RTFs. In this paper, this EVD-based PSD estimator is further analyzed and its estimation accuracy and computational complexity is analytically compared to a state-of-the-art maximum likelihood (ML)-based PSD estimator. It is shown that for perfect knowledge of the RTFs, spatial coherence matrix, and reverberant speech PSD matrix, the ML-based and EVD-based PSD estimates are both equal to the true late reverberation PSD. In addition, it is shown that for erroneous RTFs but perfect knowledge of the spatial coherence matrix and reverberant speech PSD matrix, the MLbased PSD estimate is larger than or equal to the true late reverberation PSD, whereas the EVD-based PSD estimate is obviously still equal to the true late reverberation PSD. Finally, it is shown that when modeling and estimation errors occur in all quantities, the ML-based PSD estimate is larger than or equal to the EVD-based PSD estimate. Simulation results for several realistic acoustic scenarios demonstrate the advantages of using the EVD-based PSD estimator in a multi-channel Wiener filter, yielding a significantly better performance than the ML-based PSD estimator.

show abstract

Section: Impact Of Modeling and Estimation Errors On The Ml-basedmentioning

confidence: 99%

“…In order to compare the ML-based and EVD-based PSD estimation errors ξ ml r and ξ evd r,µ , we use (42) and distinguish between the following cases:…”

Section: φ MLmentioning

confidence: 99%

Analysis of Eigenvalue Decomposition-Based Late Reverberation Power Spectral Density Estimation

Kodrasi¹,

Doclo²

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…In particular, disregarding the presence of acoustic reflections can lead to a poor performance when the direct path propagation does not dominate, i. e., for strong reverberation, or large source-array distances. Although attempts have been made to account for the presence of reverberation, e. g., [14], [15], it remains a challenge to find statistical models that are generic but not too complex, especially for the localization of multiple sources with an arbitrary array geometry. For this reason, it has become increasingly popular in recent years to employ supervised learning methods to address the problem of DOA estimation under adverse conditions.…”

Section: Introductionmentioning

confidence: 99%

Exploiting Temporal Context in CNN Based Multisource DOA Estimation

Bohlender

Spriet²,

Tirry³

et al. 2021

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Supervised learning methods are a powerful tool for direction of arrival (DOA) estimation because they can cope with adverse conditions where simplified models fail. In this work, we consider a previously proposed convolutional neural network (CNN) approach that estimates the DOAs for multiple sources from the phase spectra of the microphones. For speech, specifically, the approach was shown to work well even when trained entirely on synthetically generated data. However, as each frame is processed separately, temporal context cannot be taken into account. This prevents the exploitation of interframe signal correlations, and the fact that DOAs do not change arbitrarily over time. We therefore consider two different extensions of the CNN: the integration of a long short-term memory (LSTM) layer, or of a temporal convolutional network (TCN). In order to accommodate the incorporation of temporal context, the training data generation framework needs to be adjusted. To obtain an easily parameterizable model, we propose to employ Markov chains to realize a gradual evolution of the source activity at different times, frequencies, and directions, throughout a training sequence. A thorough evaluation demonstrates that the proposed configuration for generating training data is suitable for the tasks of single-, and multi-talker localization. In particular, we note that with temporal context, it is important to use speech, or realistic signals in general, for the sources. Experiments with recorded impulse responses and noise reveal that the CNN with the LSTM extension outperforms all other considered approaches, including the plain CNN, and the TCN extension.

show abstract

“…This enable us to derive a statistically optimal estimator for obtaining TOA estimates directly from observed microphone recordings instead of the traditional peak picking on an estimated RIR. This is inspired by the work in [13] on DOA estimation in reverberant environments. When it is desired to estimated multiple TOA's, e.g., to estimate the distance to multiple reflectors, our proposed estimator becomes computational complex due to its multidimensional nature.…”

Section: Introductionmentioning

confidence: 99%

Sound-based Distance Estimation for Indoor Navigation in the Presence of Ego Noise

Saqib¹,

Jensen²

2019

2019 27th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

An off-the-shelf drone for indoor operation would come with a variety of different sensors that are used concurrently to avoid collision with, e.g., walls, but these sensors are typically uni-directional and offers limited spatial awareness. In this paper, we propose a model-based technique for distance estimation using sound and its reflections. More specifically, the technique is estimating Time-of-Arrivals (TOAs) of the reflected sound that could infer knowledge about room geometry and help in the design of sound-based collision avoidance. Our proposed solution is thus based on probing a known sound into an environment and then estimating the TOAs of reflected sounds recorded by a single microphone. The simulated results show that our approach to estimating TOAs for reflector position estimation works up to a distance of at least 2 meters even with significant additive noise, e.g., drone ego noise.

show abstract

DOA estimation of audio sources in reverberant environments

Cited by 15 publications

References 23 publications

Analysis of Eigenvalue Decomposition-Based Late Reverberation Power Spectral Density Estimation

Analysis of Eigenvalue Decomposition-Based Late Reverberation Power Spectral Density Estimation

Exploiting Temporal Context in CNN Based Multisource DOA Estimation

Sound-based Distance Estimation for Indoor Navigation in the Presence of Ego Noise

Contact Info

Product

Resources

About