MEConformer: Highly representative embedding extractor for speaker verification via incorporating selective convolution into deep speaker encoder

计算机科学 卷积神经网络 判别式 变压器 编码器 嵌入 说话人识别 语音识别 特征提取 字错误率 模式识别(心理学) 人工智能 量子力学 操作系统 物理 电压
作者
Qiuyu Zheng,Zengzhao Chen,Zhifeng Wang,Hai Liu,Mengting Lin
出处
期刊:Expert Systems With Applications [Elsevier BV]
卷期号:244: 123004-123004 被引量:27
标识
DOI:10.1016/j.eswa.2023.123004
摘要

Transformer models have demonstrated superior performance across various domains, including computer vision, natural language processing, and speech recognition. The success of these models can be attributed to their robust parallel capacity and high computation speed, primarily reliant on the attention layer. In the domain of speaker recognition, state-of-the-art results have been achieved using convolutional neural network (CNN) architectures, particularly with speaker embeddings represented by x-vectors and r-vectors. However, existing CNN-based methods tend to focus on local features while overlooking the global dependence of voiceprint features, resulting in the loss of crucial information. Moreover, the presence of noise in audio data is an influential factor that cannot be disregarded, as it significantly impacts the extraction of discriminative speaker embeddings. To address these challenges, we propose a novel model called the Multi-Scale Expand Convolution Transformer (MEConformer). This model aims to convert variable-length audio into a fixed low-dimensional representation. The MEConformer leverages a CNN framework with expanded receptive fields to capture frame-level features effectively. Additionally, we introduce a transformer encoder that incorporates contextual dependencies, enabling the extraction of both frame-level and discourse-level feature representations. Furthermore, we present a multi-scale residual aggregation strategy, which facilitates the efficient transmission of voiceprint information across the model. By combining these innovative components, the MEConformer achieves a state-of-the-art Equal Error Rate (EER) of 3.72% on the VoxCeleb1 test set. Furthermore, it demonstrates EERs of 5.94% and 3.72% on the VoxCeleb1-H and VoxCeleb-E datasets, respectively. The code for the proposed MEConformer model will be made publicly available at https://codeocean.com/capsule/4563012/tree.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
忧心的妙之完成签到,获得积分10
1秒前
海洲完成签到 ,获得积分10
2秒前
2秒前
3秒前
酷酷的悒发布了新的文献求助10
5秒前
Ava的应助被孤独星月采纳,获得10
8秒前
言宴完成签到 ,获得积分10
8秒前
丘比特的应助被han采纳,获得10
8秒前
10秒前
10秒前
10秒前
东方元语发布了新的文献求助20
11秒前
所所的应助被科研通管家采纳,获得10
12秒前
顾矜的应助被18756828852采纳,获得10
12秒前
观山青发布了新的文献求助10
12秒前
酷波er的应助被科研通管家采纳,获得10
12秒前
xiaochaoge完成签到,获得积分10
12秒前
今后的应助被科研通管家采纳,获得10
12秒前
SciGPT的应助被科研通管家采纳,获得10
12秒前
12秒前
今后的应助被科研通管家采纳,获得10
12秒前
bkagyin的应助被科研通管家采纳,获得10
12秒前
斯文败类的应助被科研通管家采纳,获得10
12秒前
Orange的应助被科研通管家采纳,获得10
12秒前
隐形曼青的应助被科研通管家采纳,获得10
13秒前
molihuakai的应助被科研通管家采纳,获得10
13秒前
13秒前
13秒前
我下载不了论文啊完成签到,获得积分20
13秒前
123564的应助被科研通管家采纳,获得10
13秒前
13秒前
斯文败类的应助被科研通管家采纳,获得10
13秒前
传奇3的应助被科研通管家采纳,获得10
13秒前
13秒前
14秒前
科目三的应助被科研通管家采纳,获得10
14秒前
小二郎的应助被科研通管家采纳,获得10
14秒前
在水一方的应助被科研通管家采纳,获得10
14秒前
科研通AI2S的应助被科研通管家采纳,获得10
14秒前
14秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 2000
Agyptische Geschichte der 21.30. Dynastie 510
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
基于结构调控的功能化水诱导发电材料与器件研究 500
基于界面调控的高效湿气发电材料与器件研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8192723
求助须知:如何正确求助?哪些是违规求助? 9579144
关于积分的说明 21863903
捐赠科研通 7792268
什么是DOI,文献DOI怎么找? 3437013
关于科研通互助平台的介绍 2560633
邀请新用户注册赠送积分活动 2483864