Training-Free Voice Conversion with Factorized Optimal Transport
This addresses the need for efficient voice conversion in speech processing, particularly for cross-lingual applications, though it is an incremental improvement over existing methods.
The paper tackles the problem of voice conversion with minimal reference audio by introducing Factorized MKL-VC, a training-free method that replaces kNN regression with factorized optimal transport in WavLM embeddings. It achieves high-quality any-to-any cross-lingual voice conversion with only 5 seconds of reference audio, outperforming kNN-VC and matching FACodec in cross-lingual scenarios.
This paper introduces Factorized MKL-VC, a training-free modification for kNN-VC pipeline. In contrast with original pipeline, our algorithm performs high quality any-to-any cross-lingual voice conversion with only 5 second of reference audio. MKL-VC replaces kNN regression with a factorized optimal transport map in WavLM embedding subspaces, derived from Monge-Kantorovich Linear solution. Factorization addresses non-uniform variance across dimensions, ensuring effective feature transformation. Experiments on LibriSpeech and FLEURS datasets show MKL-VC significantly improves content preservation and robustness with short reference audio, outperforming kNN-VC. MKL-VC achieves performance comparable to FACodec, especially in cross-lingual voice conversion domain.