Binaural Selective Attention Model for Target Speaker Extraction
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914837947219968 |
|---|---|
| author | Meng, Hanyu Zhang, Qiquan Zhang, Xiangyu Sethu, Vidhyasaharan Ambikairajah, Eliathamby |
| author_facet | Meng, Hanyu Zhang, Qiquan Zhang, Xiangyu Sethu, Vidhyasaharan Ambikairajah, Eliathamby |
| contents | The remarkable ability of humans to selectively focus on a target speaker in cocktail party scenarios is facilitated by binaural audio processing. In this paper, we present a binaural time-domain Target Speaker Extraction model based on the Filter-and-Sum Network (FaSNet). Inspired by human selective hearing, our proposed model introduces target speaker embedding into separators using a multi-head attention-based selective attention block. We also compared two binaural interaction approaches -- the cosine similarity of time-domain signals and inter-channel correlation in learned spectral representations. Our experimental results show that our proposed model outperforms monaural configurations and state-of-the-art multi-channel target speaker extraction models, achieving best-in-class performance with 18.52 dB SI-SDR, 19.12 dB SDR, and 3.05 PESQ scores under anechoic two-speaker test configurations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_12236 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Binaural Selective Attention Model for Target Speaker Extraction Meng, Hanyu Zhang, Qiquan Zhang, Xiangyu Sethu, Vidhyasaharan Ambikairajah, Eliathamby Audio and Speech Processing Sound Signal Processing The remarkable ability of humans to selectively focus on a target speaker in cocktail party scenarios is facilitated by binaural audio processing. In this paper, we present a binaural time-domain Target Speaker Extraction model based on the Filter-and-Sum Network (FaSNet). Inspired by human selective hearing, our proposed model introduces target speaker embedding into separators using a multi-head attention-based selective attention block. We also compared two binaural interaction approaches -- the cosine similarity of time-domain signals and inter-channel correlation in learned spectral representations. Our experimental results show that our proposed model outperforms monaural configurations and state-of-the-art multi-channel target speaker extraction models, achieving best-in-class performance with 18.52 dB SI-SDR, 19.12 dB SDR, and 3.05 PESQ scores under anechoic two-speaker test configurations. |
| title | Binaural Selective Attention Model for Target Speaker Extraction |
| topic | Audio and Speech Processing Sound Signal Processing |
| url | https://arxiv.org/abs/2406.12236 |