# AAS-VC: On the Generalization Ability of Automatic Alignment Search based Non-autoregressive Sequence-to-sequence Voice Conversion

arXiv: https://arxiv.org/abs/2309.07598 · HTML: https://mlsanity.com/paper/2309.07598/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Sep 2023.

## Built on by (most cited first, 2 of 2)

- [ArVoice: A Multi-Speaker Dataset for Arabic Speech Synthesis](https://mlsanity.com/paper/2505.20506/) (2025)
- [A Pilot Study of Applying Sequence-to-Sequence Voice Conversion to Evaluate the Intelligibility of L2 Speech Using a Native Speaker's Shadowings](https://mlsanity.com/paper/2410.02239/) (2024)

## Abstract (arXiv)

Non-autoregressive (non-AR) sequence-to-seqeunce (seq2seq) models for voice conversion (VC) is attractive in its ability to effectively model the temporal structure while enjoying boosted intelligibility and fast inference thanks to non-AR modeling. However, the dependency of current non-AR seq2seq VC models on ground truth durations extracted from an external AR model greatly limits its generalization ability to smaller training datasets. In this paper, we first demonstrate the above-mentioned problem by varying the training data size. Then, we present AAS-VC, a non-AR seq2seq VC model based on automatic alignment search (AAS), which removes the dependency on external durations and serves as a proper inductive bias to provide the required generalization ability for small datasets. Experimental results show that AAS-VC can generalize better to a training dataset of only 5 minutes. We also conducted ablation studies to justify several model design choices. The audio samples and implementation are available online.