# EfficientTDNN: Efficient Architecture Search for Speaker Recognition

arXiv: https://arxiv.org/abs/2103.13581 · HTML: https://mlsanity.com/paper/2103.13581/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 55 references that are in the map.
- First posted on arXiv 25 Mar 2021.

## Builds on (most cited first, 30 of 33)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [DARTS: Differentiable Architecture Search](https://mlsanity.com/paper/1806.09055/) (2018)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [Once-for-All: Train One Network and Specialize it for Efficient Deployment](https://mlsanity.com/paper/1908.09791/) (2019)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware](https://mlsanity.com/paper/1812.00332/) (2018)
- [ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification](https://mlsanity.com/paper/2005.07143/) (2020)
- [Res2Net: A New Multi-scale Backbone Architecture](https://mlsanity.com/paper/1904.01169/) (2019)
- [Efficient Processing of Deep Neural Networks: A Tutorial and Survey](https://mlsanity.com/paper/1703.09039/) (2017)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [Single Path One-Shot Neural Architecture Search with Uniform Sampling](https://mlsanity.com/paper/1904.00420/) (2019)
- [Generalized End-to-End Loss for Speaker Verification](https://mlsanity.com/paper/1710.10467/) (2017)
- [In defence of metric learning for speaker recognition](https://mlsanity.com/paper/2003.11982/) (2020)
- [Utterance-level Aggregation For Speaker Recognition In The Wild](https://mlsanity.com/paper/1902.10107/) (2019)
- [Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap](https://mlsanity.com/paper/2008.01475/) (2020)
- [Semi-Supervised Neural Architecture Search](https://mlsanity.com/paper/2002.10389/) (2020)
- [Clova Baseline System for the VoxCeleb Speaker Recognition Challenge 2020](https://mlsanity.com/paper/2009.14153/) (2020)
- [Self-attention encoding and pooling for speaker recognition](https://mlsanity.com/paper/2008.01077/) (2020)
- [LightSpeech: Lightweight and Fast Text to Speech with Neural Architecture Search](https://mlsanity.com/paper/2102.04040/) (2021)
- [Margin Matters: Towards More Discriminative Deep Neural Network Embeddings for Speaker Recognition](https://mlsanity.com/paper/1906.07317/) (2019)
- [Learning Autoencoders with Relational Regularization](https://mlsanity.com/paper/2002.02913/) (2020)
- [How to Train Your Super-Net: An Analysis of Training Heuristics in Weight-Sharing NAS](https://mlsanity.com/paper/2003.04276/) (2020)
- [AutoSpeech: Neural Architecture Search for Speaker Recognition](https://mlsanity.com/paper/2005.03215/) (2020)
- [DARTS-ASR: Differentiable Architecture Search for Multilingual Speech Recognition and Adaptation](https://mlsanity.com/paper/2005.07029/) (2020)
- [VoxSRC 2019: The first VoxCeleb Speaker Recognition Challenge](https://mlsanity.com/paper/1912.02522/) (2019)
- [Unified Hypersphere Embedding for Speaker Recognition](https://mlsanity.com/paper/1807.08312/) (2018)

## Built on by (most cited first, 3 of 3)

- [LightHuBERT: Lightweight and Configurable Speech Representation Learning with Once-for-All Hidden-Unit BERT](https://mlsanity.com/paper/2203.15610/) (2022)
- [Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models](https://mlsanity.com/paper/2503.06027/) (2025)
- [Gaussian Flow Bridges for Audio Domain Transfer with Unpaired Data](https://mlsanity.com/paper/2405.19497/) (2024)

## Abstract (arXiv)

Convolutional neural networks (CNNs), such as the time-delay neural network (TDNN), have shown their remarkable capability in learning speaker embedding. However, they meanwhile bring a huge computational cost in storage size, processing, and memory. Discovering the specialized CNN that meets a specific constraint requires a substantial effort of human experts. Compared with hand-designed approaches, neural architecture search (NAS) appears as a practical technique in automating the manual architecture design process and has attracted increasing interest in spoken language processing tasks such as speaker recognition. In this paper, we propose EfficientTDNN, an efficient architecture search framework consisting of a TDNN-based supernet and a TDNN-NAS algorithm. The proposed supernet introduces temporal convolution of different ranges of the receptive field and feature aggregation of various resolutions from different layers to TDNN. On top of it, the TDNN-NAS algorithm quickly searches for the desired TDNN architecture via weight-sharing subnets, which surprisingly reduces computation while handling the vast number of devices with various resources requirements. Experimental results on the VoxCeleb dataset show the proposed EfficientTDNN enables approximate 1013 architectures concerning depth, kernel, and width. Considering different computation constraints, it achieves a 2.20% equal error rate (EER) with 204M multiply-accumulate operations (MACs), 1.41% EER with 571M MACs as well as 0.94% EER with 1.45G MACs. Comprehensive investigations suggest that the trained supernet generalizes subnets not sampled during training and obtains a favorable trade-off between accuracy and efficiency.