Spatial Pyramid Encoding with Convex Length Normalization for Text-Independent Speaker Verification

Abstract

In this paper, we propose a new pooling method called spatial pyramid encoding (SPE) to generate speaker embeddings for text-independent speaker verification. We first partition the output feature maps from a deep residual network (ResNet) into increasingly fine sub-regions and extract speaker embeddings from each sub-region through a learnable dictionary encoding layer. These embeddings are concatenated to obtain the final speaker representation. The SPE layer not only generates a fixed-dimensional speaker embedding for a variable-length speech segment, but also aggregates the information of feature distribution from multi-level temporal bins. Furthermore, we apply deep length normalization by augmenting the loss function with ring loss. By applying ring loss, the network gradually learns to normalize the speaker embeddings using model weights themselves while preserving convexity, leading to more robust speaker embeddings. Experiments on the VoxCeleb1 dataset show that the proposed system using the SPE layer and ring loss-based deep length normalization outperforms both i-vector and d-vector baselines.

arXiv comment5 pages, 2 figures, Interspeech 2019

References in corpus (6)

• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• VoxCeleb2: Deep Speaker Recognition
• VoxCeleb: a large-scale speaker identification dataset
• Attentive Statistics Pooling for Deep Speaker Embedding
• Deep Speaker: an End-to-End Neural Speaker Embedding System
• L2-constrained Softmax Loss for Discriminative Face Verification

Cited by in corpus (15)

• Improving Multi-Scale Aggregation Using Feature Pyramid Module for Robust Speaker Verification of Variable-Duration Utterances
• Deep MOS Predictor for Synthetic Speech Using Cluster-Based Modeling
• A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments
• Meta-Learning for Short Utterance Speaker Recognition with Imbalance Length Pairs
• Speaker Recognition Based on Deep Learning: An Overview
• Toward Improving Synthetic Audio Spoofing Detection Robustness via Meta-Learning and Disentangled Training With Adversarial Examples
• Zero-Shot Multi-Speaker Text-To-Speech with State-of-the-art Neural Speaker Embeddings
• Improved RawNet with Feature Map Scaling for Text-independent Speaker Verification using Raw Waveforms
• Self-Adaptive Soft Voice Activity Detection using Deep Neural Networks for Robust Speaker Verification
• Speaker Characterization by means of Attention Pooling
• MCSAE: Masked Cross Self-Attentive Encoding for Speaker Embedding
• Speaker Representation Learning using Global Context Guided Channel and Time-Frequency Transformations
• Multi-Task Network for Noise-Robust Keyword Spotting and Speaker Verification using CTC-based Soft VAD and Global Query Attention
• Bidirectional Multiscale Feature Aggregation for Speaker Verification
• Supervised attention for speaker recognition