通过插值说话人嵌入生成新语音数据，提升验证模型性能。
Interpolating Speaker Identities in Embedding Space for Data Expansion
- 在预训练嵌入空间中对相邻说话人进行球面插值生成新身份
- 在说话人验证任务上实现3.06%~5.24%相对性能提升
- 适用于说话人验证与性别分类，可与其他增强方法兼容
深度学习驱动的说话人验证系统性能依赖于大规模、多样化的说话人数据。然而，获取更多说话人数据成本高、难度大，且受隐私限制。为此，我们提出INSIDE（Interpolating Speaker Identities in Embedding Space），一种新型数据扩展方法：通过在预训练说话人嵌入空间中对相近说话人嵌入进行球面线性插值，生成中间嵌入，并输入文语转换系统生成对应语音波形。将生成数据与原始数据结合训练下游模型。实验表明，使用INSIDE扩展数据训练的模型优于仅使用真实数据的模型，在说话人验证任务上实现3.06%至5.24%的相对性能提升。尽管主要面向说话人验证，该方法在性别分类任务上也取得13.44%的相对改进。此外，INSIDE可与其他增强技术兼容，能作为灵活、可扩展的训练流程补充。
原文摘要 · Abstract (English)
The success of deep learning-based speaker verification systems is largely attributed to access to large-scale and diverse speaker identity data. However, collecting data from more identities is expensive, challenging, and often limited by privacy concerns. To address this limitation, we propose INSIDE (Interpolating Speaker Identities in Embedding Space), a novel data expansion method that synthesizes new speaker identities by interpolating between existing speaker embeddings. Specifically, we select pairs of nearby speaker embeddings from a pretrained speaker embedding space and compute intermediate embeddings using spherical linear interpolation. These interpolated embeddings are then fed to a text-to-speech system to generate corresponding speech waveforms. The resulting data is combined with the original dataset to train downstream models. Experiments show that models trained with INSIDE-expanded data outperform those trained only on real data, achieving 3.06\% to 5.24\% relative improvements. While INSIDE is primarily designed for speaker verification, we also validate its effectiveness on gender classification, where it yields a 13.44\% relative improvement. Moreover, INSIDE is compatible with other augmentation techniques and can serve as a flexible, scalable addition to existing training pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。