Performance measurement in blind audio source separation

By Emmanuel Vincent, Rémi Gribonval, Cédric Févotte. 2006

3,197 citations recorded by the source corpus.

Subfields: Computational Mechanics, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Speech & Audio

In Signal Processing

Citations in the corpus

Differentiable Model Compression via Pseudo Quantization Noise (2021, methodology): “The model is trained on the standard MusDB benchmark (Rafii et al. 2017), for 180 epochs, and evaluated with the Signal-To-Distortion Ratio (SDR) metric (Vincent, Gribonval, and Févotte 2006).”
Music Source Separation in the Waveform Domain (2019, background): “His framework for auditory scene analysis [Bregman, 1990] spawned its computational counterpart, trying to reproduce or model accomplishments of the brains with algorithmic means [Wang and Brown, 2006], in particular regarding source separation capabilities.”
Audio-Visual Scene Analysis with Self-Supervised Multisensory Features (2018, methodology): “While these methods are not applicable to on/off-screen separation, we modified our model to have it separate audio using an extra permutation invariant loss (Equation 4) and then compared the methods using blind separation metrics [68]: signal-to-distortion ratio (SDR), signal-to-interference ratio”
Multitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural Networks (2017, methodology): “The Ideal Ratio Mask (IRM) [27] for each source is deﬁned as When the phase of Y is used for reconstruction, the IRM achieves the highest Signal to Distortion Ratio (SDR) [44], when all sources have the same phase, (which is an in-valid assumption in general).”
TaSNet: Time-Domain Audio Separation Network for Real-Time, Single-Channel Speech Separation (2017, background): “fined as scale-invariant SNR (SI-SNR), and the standard SDR proposed in [25].”
Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures (2014, background): “Our evaluation measure is source-to-distortion ratio (SDR) [28].”
SDR – Half-baked or Well Done? (2018, methodology): “Note that the original BSS eval paper [17] also considered time-varying gains and time-varying filters as allowed deformations.”
Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation (2018, methodology): “The signal-to-distortion (SDR) metric is commonly used to evaluate source separation performance [25].”
Permutation invariant training of deep models for speaker-independent multi-talker speech separation (2016, methodology): “First, better SDR improvement can be achieved by estimating less output frames.”
The Sound of Pixels (2018, background): “While auditory scene analysis [1] is widely studied in the fields of environmental sound recognition [2,3] and source separation [4,5,6,7,8,9], the natural synchronization between vision and sound can provide a rich supervisory signal for grounding sounds in vision [10,11,12].”
Dual-Path RNN: Efficient Long Sequence Modeling for Time-Domain Single-Channel Speech Separation (2019, methodology): “The degree of improvement in the signal fidelity is measured by signal-to-distortion ratio improvement (SDRi) [30] as well as SI-SNR improvement (SISNRi).”
End-to-End Speech Separation with Unfolded Iterative Phase Reconstruction (2018, methodology): “We report the performance using scale-invariant SDR (SISDR) [1, 2, 5, 39], as well as the SDR metric computed using the bss eval sources software [40] because it is used by other groups.”
Improving singing voice separation using Deep U-Net and Wave-U-Net with data augmentation (2019, methodology): “We compute the following three metrics [22]: Source-to-Interference Ratios (SIR), Source-to-Artifact Ratios (SAR) and Source-to-Distortion Ratios (SDR) and report the median over the test database.”
A Deep Ensemble Learning Method for Monaural Speech Separation (2016, methodology): “3) Evaluation Metrics: Besides STOI, we used the source to distortion ratio (SDR) [28], a metric similar to SNR for evaluating the quality of separation.”
Phase-aware Speech Enhancement with Deep Complex U-Net (2019, methodology): “Taking this as an advantage, we introduce a novel loss function which directly optimizes source-to-distortion ratio (SDR) (Vincent et al., 2006), a quantitative evaluation measure widely used in many source separation tasks.”
Stable and fast update rules for independent vector analysis based on auxiliary function technique (2011, neutral): “(37)”
Semantic Communication Systems for Speech Transmission (2021, methodology): “Hence, the signal-to-distortion ration (SDR) [32] is employed to measure the L2 error between s and ŝ, which is one of the commonly used metric for speech transmission and can be expressed as”