BSS Eval or Peass? Predicting the Perception of Singing-Voice Separation

Ward, Dominic; Wierstorf, Hagen; Mason, Russell; Grais, Emad M.; Plumbley, Mark D.

doi:10.1109/icassp.2018.8462194

Cited by 10 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second step provides performance criteria from the computation of energy ratios related to the previous four terms: source to distortion ratio (SDR), source to interferences ratio (SIR), sources to noise ratio and sources to artefacts ratio (SAR). Although a reasonable correlation was found between SIR and human ratings of interference [268], other experiments [27], [268] showed that energy-based measures are not ideal for determining perceptual sound quality for SS algorithms.…”

Section: B Objective Measuresmentioning

confidence: 91%

Deep-learning-based audio-visual speech enhancement in presence of Lombard effect

Michelsanti

Sigurðsson

Jensen

2019

Speech Communication

View full text Add to dashboard Cite

Keywords:Lombard effect audio-visual speech enhancement deep learning speech quality speech intelligibility A B S T R A C T When speaking in presence of background noise, humans reflexively change their way of speaking in order to improve the intelligibility of their speech. This reflex is known as Lombard effect. Collecting speech in Lombard conditions is usually hard and costly. For this reason, speech enhancement systems are generally trained and evaluated on speech recorded in quiet to which noise is artificially added. Since these systems are often used in situations where Lombard speech occurs, in this work we perform an analysis of the impact that Lombard effect has on audio, visual and audio-visual speech enhancement, focusing on deep-learning-based systems, since they represent the current state of the art in the field.We conduct several experiments using an audio-visual Lombard speech corpus consisting of utterances spoken by 54 different talkers. The results show that training deep-learning-based models with Lombard speech is beneficial in terms of both estimated speech quality and estimated speech intelligibility at low signal to noise ratios, where the visual modality can play an important role in acoustically challenging situations. We also find that a performance difference between genders exists due to the distinct Lombard speech exhibited by males and females, and we analyse it in relation with acoustic and visual features. Furthermore, listening tests conducted with audio-visual stimuli show that the speech quality of the signals processed with systems trained using Lombard speech is statistically significantly better than the one obtained using systems trained with non-Lombard speech at a signal to noise ratio of −5 dB. Regarding speech intelligibility, we find a general tendency of the benefit in training the systems with Lombard speech.

show abstract

Section: B Objective Measuresmentioning

confidence: 91%

Deep-learning-based audio-visual speech enhancement in presence of Lombard effect

Michelsanti

Sigurðsson

Jensen

2019

Speech Communication

View full text Add to dashboard Cite

show abstract

“…Also, the separation algorithm should preserve speech quality. This aspect is evaluated using perceptual evaluation of speech quality (PESQ) [40], which maintains a good correlation with speech intelligibility measures even at lower SNR [45]. Another important challenge of SS systems is speech intelligibility that should not be affected by the separation process.…”

Section: Discussionmentioning

confidence: 99%

“…Another important challenge of SS systems is speech intelligibility that should not be affected by the separation process. This aspect is evaluated in terms of STOI [41], which is characterised with a high correlation with speech intelligibility compared to other existing objective intelligibility models [45]. The proposed approach is evaluated and compared to other states-of-the-art approaches.…”

Section: Experiments 2: Evaluation Of the Whole Separation Systemmentioning

confidence: 99%

Phase‐aware subspace decomposition for single channel speech separation

Wiem

Anouar

Bouzid

2020

IET signal process.

View full text Add to dashboard Cite

Single channel speech separation (SCSS) is often required as post-processing in several applications that facilitate automatic human-to-human or human-to-machine communication in challenging acoustic environments such as voice command for smart homes or robotics. The proposed SCSS system, that the authors call phase-aware subspace decomposition (PASD), relies on subspace decomposition for speech separation followed by a phase-aware mask for final subspace recovery. In fact, the proposed approach decomposes the mixture into a sparse and low-rank subspace in the frequency domain by rank minimising that relies on iterative decomposition using adaptive thresholding in each iteration to achieve soft estimation and considers phase-information for reconstruction. Separation results are reported in terms of both intrusive and non-intrusive metrics using realistic recordings corrupted with real-life noises. As speech separation systems are expected to have maximal interference rejection without speech distortion, we also evaluate the proposed system by recognising speech from a target speaker in the presence of either concurrent speech or noise. Recognition results show that separated signals are of high intelligibility so that they can be exploited by other automatic applications. The extensive evaluation under different test scenarios proves that PASD consistently improves the quality of the separated signals, compared to other benchmark approaches.

show abstract

“…As previously mentioned, a unified, robust and perceptually valid MSS quality evaluation procedure does not yet exist. Even while new alternatives for evaluation have been explored in recent years [30], listening tests remain the only reliable quality evaluation method to date.…”

Section: Future Research Directionsmentioning

confidence: 99%