arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.07770eess.AScs.AI
收藏

层感知TDNN:利用预训练模型的多层特征进行说话人识别

Layer-aware TDNN: Speaker Recognition Using Multi-Layer Features from Pre-Trained Models

Department of Industrial and Management Engineering(工业与管理工程系)

Jin Sob Kim, Hyun Joon Park, Wooseok Shin, Juan Yun, Sung Won Han

AI总结:

本文提出L-TDNN,通过层感知处理提升说话人识别性能,实现低错误率和高效推理。

摘要:

最近在Transformer上的自监督学习(SSL)进展显著提高了说话人验证(SV)的性能,通过提供通用语音表示。然而,现有方法未能充分利用SSL编码器的多层特性。为解决这一限制,我们提出了层感知时间延迟神经网络(L-TDNN),该网络直接对预训练模型的层-wise隐藏状态输出进行层/帧级处理,提取固定大小的说话人向量。L-TDNN包括一个层感知卷积网络、帧适应的层聚合和注意力统计池化,明确建模了之前被忽视的层维度的识别和处理。我们评估了L-TDNN在多个语音SSL Transformer和多样化的语音-说话人语料库上,与其他方法相比利用预训练编码器的性能。L-TDNN在多个实验中均表现出稳健的验证性能,实现了最低的错误率。同时,它在模型紧凑性方面也表现出色,并且推理效率与现有系统相当。这些结果突显了所提出层感知处理方法的优势。未来的工作包括探索与SSL前端的联合训练以及纳入分数校准以进一步提升最先进的验证性能。

英文摘要:

Recent advances in self-supervised learning (SSL) on Transformers have significantly improved speaker verification (SV) by providing domain-general speech representations. However, existing approaches have underutilized the multi-layered nature of SSL encoders. To address this limitation, we propose the layer-aware time-delay neural network (L-TDNN), which directly performs layer/frame-wise processing on the layer-wise hidden state outputs from pre-trained models, extracting fixed-size speaker vectors. L-TDNN comprises a layer-aware convolutional network, a frame-adaptive layer aggregation, and attentive statistic pooling, explicitly modeling of the recognition and processing of previously overlooked layer dimension. We evaluated L-TDNN across multiple speech SSL Transformers and diverse speech-speaker corpora against other approaches for leveraging pre-trained encoders. L-TDNN consistently demonstrated robust verification performance, achieving the lowest error rates throughout the experiments. Concurrently, it stood out in terms of model compactness and exhibited inference efficiency comparable to the existing systems. These results highlight the advantages derived from the proposed layer-aware processing approach. Future work includes exploring joint training with SSL frontends and the incorporation of score calibration to further enhance state-of-the-art verification performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑