Multi-label Training For Text-independent Speaker Identification

Abstract

In this paper, we propose a novel strategy for text-independent speaker identification system: Multi-Label Training (MLT). Instead of the commonly used one-to-one correspondence between the speech and the speaker label, we divide all the speeches of each speaker into several subgroups, with each subgroup assigned a different set of labels. During the identification process, a specific speaker is identified as long as the predicted label is the same as one of his/her corresponding labels. We found that this method can force the model to distinguish the data more accurately, and somehow takes advantages of ensemble learning, while avoiding the significant increase of computation and storage burden. In the experiments, we found that not only in clean conditions, but also in noisy conditions with speech enhancement, Multi-Label Training can still achieve better identification performance than commom methods. It should be noted that the proposed strategy can be easily applied to almost all

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyxue2022multi

Related papers

• A Purely End-to-end System For Multi-speaker Speech Recognition (2018)12.25
• PCA/LDA Approach For Text-independent Speaker Recognition (2016)5.24
• Multiple Choice Learning For Efficient Speech Separation With Many Speakers (2024)2.26
• Multi-task Metric Learning For Text-independent Speaker Verification (2020)0.00
• TIMIT Speaker Profiling: A Comparison Of Multi-task Learning And Single-task Learning Approaches (2024)0.00
• Multiobjective Optimization Training Of PLDA For Speaker Verification (2018)2.26
• Single-channel Multi-talker Speech Recognition With Permutation Invariant Training (2017)12.10
• Recognizing Multi-talker Speech With Permutation Invariant Training (2017)12.81