FlowTSE: Target Speaker Extraction with Flow Matching
This work addresses the problem of isolating a specific speaker's speech from mixtures for applications like speech enhancement, presenting an incremental improvement in generative methods for TSE.
The paper tackled target speaker extraction by proposing FlowTSE, a generative method based on conditional flow matching, which matches or outperforms strong baselines on standard benchmarks.
Target speaker extraction (TSE) aims to isolate a specific speaker's speech from a mixture using speaker enrollment as a reference. While most existing approaches are discriminative, recent generative methods for TSE achieve strong results. However, generative methods for TSE remain underexplored, with most existing approaches relying on complex pipelines and pretrained components, leading to computational overhead. In this work, we present FlowTSE, a simple yet effective TSE approach based on conditional flow matching. Our model receives an enrollment audio sample and a mixed speech signal, both represented as mel-spectrograms, with the objective of extracting the target speaker's clean speech. Furthermore, for tasks where phase reconstruction is crucial, we propose a novel vocoder conditioned on the complex STFT of the mixed signal, enabling improved phase estimation. Experimental results on standard TSE benchmarks show that FlowTSE matches or outperforms strong baselines.