DeepAI AI Chat
Log In Sign Up

H-VECTORS: Utterance-level Speaker Embedding Using A Hierarchical Attention Model

by   Yanpei Shi, et al.

In this paper, a hierarchical attention network to generate utterance-level embeddings (H-vectors) for speaker identification is proposed. Since different parts of an utterance may have different contributions to speaker identities, the use of hierarchical structure aims to learn speaker related information locally and globally. In the proposed approach, frame-level encoder and attention are applied on segments of an input utterance and generate individual segment vectors. Then, segment level attention is applied on the segment vectors to construct an utterance representation. To evaluate the effectiveness of the proposed approach, NIST SRE 2008 Part1 dataset is used for training, and two datasets, Switchboard Cellular part1 and CallHome American English Speech, are used to evaluate the quality of extracted utterance embeddings on speaker identification and verification tasks. In comparison with two baselines, X-vector, X-vector+Attention, the obtained results show that H-vectors can achieve a significantly better performance. Furthermore, the extracted utterance-level embeddings are more discriminative than the two baselines when mapped into a 2D space using t-SNE.


page 1

page 2

page 3

page 4


Weakly Supervised Training of Hierarchical Attention Networks for Speaker Identification

Identifying multiple speakers without knowing where a speaker's voice is...

Effective Incorporation of Speaker Information in Utterance Encoding in Dialog

In dialog studies, we often encode a dialog using a hierarchical encoder...

Probabilistic embeddings for speaker diarization

Speaker embeddings (x-vectors) extracted from very short segments of spe...

Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal

Speech signal is constituted and contributed by various informative fact...

T-vectors: Weakly Supervised Speaker Identification Using Hierarchical Transformer Model

Identifying multiple speakers without knowing where a speaker's voice is...

Speaker-Utterance Dual Attention for Speaker and Utterance Verification

In this paper, we study a novel technique that exploits the interaction ...

Speaker Diarization Using Stereo Audio Channels: Preliminary Study on Utterance Clustering

Speaker diarization is one of the actively researched topics in audio si...