Compositional Embedding Models for Speaker Identification and Diarization with Simultaneous Speech From 2+ Speakers
Compositional Embedding Models for Speaker Identification and Diarization with Simultaneous Speech From 2+ Speakers
复制标题
DOI：
10.1109/icassp39728.2021.9413752
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Zeqian Li;J. Whitehill
中科院分区：
文献类型：
--
作者：
Zeqian Li;J. Whitehill
我们提出了一种新的说话人划分方法，可以处理2人以上的重叠语音。我们的方法基于组合嵌入[1]：与标准的说话者嵌入方法（如x-vector[2]）一样，组合嵌入模型包含一个函数f，用于将不同说话者的语音分离开来。此外，它们还包括一个组合函数g，用于计算嵌入空间中的集合并运算，从而推断输入音频中的扬声器集合。在使用合成的LibriSpeech数据进行多人说话人识别的实验中，该方法优于传统的只训练分离单个说话人（而不是说话人集）的嵌入方法。在AMI耳机混合语料库上的扬声器diarization实验中，我们达到了最先进的准确率（DER=22.93%），略好于之前的最佳结果（[3]的23.82%）。
We propose a new method for speaker diarization that can handle overlapping speech with 2+ people. Our method is based on compositional embeddings [1]: Like standard speaker embedding methods such as x-vector [2], compositional embedding models contain a function f that separates speech from different speakers. In addition, they include a composition function g to compute set-union operations in the embedding space so as to infer the set of speakers within the input audio. In an experiment on multi-person speaker identification using synthesized LibriSpeech data, the proposed method outperforms traditional embedding methods that are only trained to separate single speakers (not speaker sets). In a speaker diarization experiment on the AMI Headset Mix corpus, we achieve state-of-the-art accuracy (DER=22.93%), slightly better than the previous best result (23.82% from [3]).