Location-based training for multi-channel talker-independent speaker separation
This work addresses a key bottleneck in speaker separation for audio processing applications, offering a more efficient and effective method, though it is incremental as it builds on existing spatial information approaches.
The paper tackles the permutation ambiguity problem in multi-channel speaker separation by proposing location-based training (LBT), which assigns speakers based on spatial locations like azimuth angles and distances. Results show that LBT consistently outperforms permutation-invariant training (PIT) for two- and three-speaker mixtures, with dynamic selection further improving performance, and it scales linearly with speaker count compared to PIT's factorial complexity.
Permutation-invariant training (PIT) is a dominant approach for addressing the permutation ambiguity problem in talker-independent speaker separation. Leveraging spatial information afforded by microphone arrays, we propose a new training approach to resolving permutation ambiguities for multi-channel speaker separation. The proposed approach, named location-based training (LBT), assigns speakers on the basis of their spatial locations. This training strategy is easy to apply, and organizes speakers according to their positions in physical space. Specifically, this study investigates azimuth angles and source distances for location-based training. Evaluation results on separating two- and three-speaker mixtures show that azimuth-based training consistently outperforms PIT, and distance-based training further improves the separation performance when speaker azimuths are close. Furthermore, we dynamically select azimuth-based or distance-based training by estimating the azimuths of separated speakers, which further improves separation performance. LBT has a linear training complexity with respect to the number of speakers, as opposed to the factorial complexity of PIT. We further demonstrate the effectiveness of LBT for the separation of four and five concurrent speakers.