# SA-SDR: A novel loss function for separation of meeting style data

arXiv: https://arxiv.org/abs/2110.15581 · HTML: https://mlsanity.com/paper/2110.15581/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Oct 2021.

## Built on by (most cited first, 4 of 4)

- [DeFT-Mamba: Universal Multichannel Sound Separation and Polyphonic Audio Classification](https://mlsanity.com/paper/2409.12413/) (2024)
- [Plug-and-Play Co-Occurring Face Attention for Robust Audio-Visual Speaker Extraction](https://mlsanity.com/paper/2505.20635/) (2025)
- [Combining TF-GridNet and Mixture Encoder for Continuous Speech Separation for Meeting Transcription](https://mlsanity.com/paper/2309.08454/) (2023)
- [MeCo: One-Step MeanFlow-based Corrector for Multi-Channel Speech Separation](https://mlsanity.com/paper/2606.09677/) (2026)

## Abstract (arXiv)

Many state-of-the-art neural network-based source separation systems use the averaged Signal-to-Distortion Ratio (SDR) as a training objective function. The basic SDR is, however, undefined if the network reconstructs the reference signal perfectly or if the reference signal contains silence, e.g., when a two-output separator processes a single-speaker recording. Many modifications to the plain SDR have been proposed that trade-off between making the loss more robust and distorting its value. We propose to switch from a mean over the SDRs of each individual output channel to a global SDR over all output channels at the same time, which we call source-aggregated SDR (SA-SDR). This makes the loss robust against silence and perfect reconstruction as long as at least one reference signal is not silent. We experimentally show that our proposed SA-SDR is more stable and preferable over other well-known modifications when processing meeting-style data that typically contains many silent or single-speaker regions.