WAND: Windowed Attention and Knowledge Distillation for Efficient Autoregressive Text-to-Speech Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910118125240320 |
|---|---|
| author | Lee, Hanna Nguyen, Tan Dat Kang, Jaehoon Shim, Kyuhong |
| author_facet | Lee, Hanna Nguyen, Tan Dat Kang, Jaehoon Shim, Kyuhong |
| contents | Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed Attention and Knowledge Distillation, a framework that adapts pretrained AR-TTS models to operate with constant computational and memory complexity. WAND separates the attention mechanism into two: persistent global attention over conditioning tokens and local sliding-window attention over generated tokens. To stabilize fine-tuning, we employ a curriculum learning strategy that progressively tightens the attention window. We further utilize knowledge distillation from a full-attention teacher to recover high-fidelity synthesis quality with high data efficiency. Evaluated on three modern AR-TTS models, WAND preserves the original quality while achieving up to 66.2% KV cache memory reduction and length-invariant, near-constant per-step latency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08558 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | WAND: Windowed Attention and Knowledge Distillation for Efficient Autoregressive Text-to-Speech Models Lee, Hanna Nguyen, Tan Dat Kang, Jaehoon Shim, Kyuhong Computation and Language Artificial Intelligence Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed Attention and Knowledge Distillation, a framework that adapts pretrained AR-TTS models to operate with constant computational and memory complexity. WAND separates the attention mechanism into two: persistent global attention over conditioning tokens and local sliding-window attention over generated tokens. To stabilize fine-tuning, we employ a curriculum learning strategy that progressively tightens the attention window. We further utilize knowledge distillation from a full-attention teacher to recover high-fidelity synthesis quality with high data efficiency. Evaluated on three modern AR-TTS models, WAND preserves the original quality while achieving up to 66.2% KV cache memory reduction and length-invariant, near-constant per-step latency. |
| title | WAND: Windowed Attention and Knowledge Distillation for Efficient Autoregressive Text-to-Speech Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2604.08558 |