论文标题
目标扬声器语音活动检测：晚餐聚会场景中多演讲者诊断的新方法
Target-Speaker Voice Activity Detection: a Novel Approach for Multi-Speaker Diarization in a Dinner Party Scenario
论文作者
论文摘要
现实生活中的扬声器诊断是一个极具挑战性的问题。在这种情况下，基于聚类的诊断方法的广泛性效果相当低，这主要是由于处理重叠语音的能力有限。我们提出了一种新颖的宣传者语音活动检测（TS-VAD）方法，该方法直接预测了每个说话者在每个时间范围内的活动。 TS-VAD模型采用常规的语音特征（例如MFCC）以及每个说话者的I-向量作为输入。一组二元分类输出层会产生每个说话者的活动。可以迭代地估算I-向量，从强大的基于聚类的诊断开始。我们还使用从单渠道TS-VAD模型中提取的隐藏表示形式上使用简单的注意机制将TS-VAD方法扩展到多微型频道。此外，研究了预测的说话者活动概率的后处理策略。 Chime-6未分段数据上的实验表明，TS-VAD达到最先进的结果，其结果的表现优于基线X-vector系统，超过30％的诊断错误率（DER）ABS。
Speaker diarization for real-life scenarios is an extremely challenging problem. Widely used clustering-based diarization approaches perform rather poorly in such conditions, mainly due to the limited ability to handle overlapping speech. We propose a novel Target-Speaker Voice Activity Detection (TS-VAD) approach, which directly predicts an activity of each speaker on each time frame. TS-VAD model takes conventional speech features (e.g., MFCC) along with i-vectors for each speaker as inputs. A set of binary classification output layers produces activities of each speaker. I-vectors can be estimated iteratively, starting with a strong clustering-based diarization. We also extend the TS-VAD approach to the multi-microphone case using a simple attention mechanism on top of hidden representations extracted from the single-channel TS-VAD model. Moreover, post-processing strategies for the predicted speaker activity probabilities are investigated. Experiments on the CHiME-6 unsegmented data show that TS-VAD achieves state-of-the-art results outperforming the baseline x-vector-based system by more than 30% Diarization Error Rate (DER) abs.
