Filterbank Design For End-to-end Speech Separation

Abstract

Single-channel speech separation has recently made great progress thanks to learned filterbanks as used in ConvTasNet. In parallel, parameterized filterbanks have been proposed for speaker recognition where only center frequencies and bandwidths are learned. In this work, we extend real-valued learned and parameterized filterbanks into complex-valued analytic filterbanks and define a set of corresponding representations and masking strategies. We evaluate these filterbanks on a newly released noisy speech separation dataset (WHAM). The results show that the proposed analytic learned filterbank consistently outperforms the real-valued filterbank of ConvTasNet. Also, we validate the use of parameterized filterbanks and show that complex-valued representations and masks are beneficial in all conditions. Finally, we show that the STFT achieves its best performance for 2ms windows.

Stats

• citations41
• S2 citations—
• github stars0
• HF likes0
• heat score12.17
• arxiv keypariente2019filterbank

Related papers

• End-to-end Multi-channel Speech Separation (2019)0.00
• Dual-path Filter Network: Speaker-aware Modeling For Speech Separation (2021)3.58
• Temporal-spatial Neural Filter: Direction Informed End-to-end Multi-channel Target Speech Separation (2020)0.00
• Conv-tasnet: Surpassing Ideal Time-frequency Magnitude Masking For Speech Separation (2018)24.08
• Exploring The Integration Of Speech Separation And Recognition With Self-supervised Learning Representation (2023)6.34
• Voicefilter: Targeted Voice Separation By Speaker-conditioned Spectrogram Masking (2018)17.48
• Implicit Filter-and-sum Network For Multi-channel Speech Separation (2020)8.35
• End-to-end Networks For Supervised Single-channel Speech Separation (2018)0.00