SegINR: Segment-wise Implicit Neural Representation for Sequence Alignment in Neural Text-to-Speech
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910658335866880 |
|---|---|
| author | Kim, Minchan Jeong, Myeonghun Lee, Joun Yeop Kim, Nam Soo |
| author_facet | Kim, Minchan Jeong, Myeonghun Lee, Joun Yeop Kim, Nam Soo |
| contents | We present SegINR, a novel approach to neural Text-to-Speech (TTS) that addresses sequence alignment without relying on an auxiliary duration predictor and complex autoregressive (AR) or non-autoregressive (NAR) frame-level sequence modeling. SegINR simplifies the process by converting text sequences directly into frame-level features. It leverages an optimal text encoder to extract embeddings, transforming each into a segment of frame-level features using a conditional implicit neural representation (INR). This method, named segment-wise INR (SegINR), models temporal dynamics within each segment and autonomously defines segment boundaries, reducing computational costs. We integrate SegINR into a two-stage TTS framework, using it for semantic token prediction. Our experiments in zero-shot adaptive TTS scenarios demonstrate that SegINR outperforms conventional methods in speech quality with computational efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_04690 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SegINR: Segment-wise Implicit Neural Representation for Sequence Alignment in Neural Text-to-Speech Kim, Minchan Jeong, Myeonghun Lee, Joun Yeop Kim, Nam Soo Audio and Speech Processing Machine Learning We present SegINR, a novel approach to neural Text-to-Speech (TTS) that addresses sequence alignment without relying on an auxiliary duration predictor and complex autoregressive (AR) or non-autoregressive (NAR) frame-level sequence modeling. SegINR simplifies the process by converting text sequences directly into frame-level features. It leverages an optimal text encoder to extract embeddings, transforming each into a segment of frame-level features using a conditional implicit neural representation (INR). This method, named segment-wise INR (SegINR), models temporal dynamics within each segment and autonomously defines segment boundaries, reducing computational costs. We integrate SegINR into a two-stage TTS framework, using it for semantic token prediction. Our experiments in zero-shot adaptive TTS scenarios demonstrate that SegINR outperforms conventional methods in speech quality with computational efficiency. |
| title | SegINR: Segment-wise Implicit Neural Representation for Sequence Alignment in Neural Text-to-Speech |
| topic | Audio and Speech Processing Machine Learning |
| url | https://arxiv.org/abs/2410.04690 |