# Optimizing Multi-Taper Features for Deep Speaker Verification

arXiv: https://arxiv.org/abs/2110.10983 · HTML: https://mlsanity.com/paper/2110.10983/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 33 references that are in the map.
- First posted on arXiv 21 Oct 2021.

## Builds on (most cited first, 11 of 11)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification](https://mlsanity.com/paper/2005.07143/) (2020)
- [Speaker Recognition from Raw Waveform with SincNet](https://mlsanity.com/paper/1808.00158/) (2018)
- [Attentive Statistics Pooling for Deep Speaker Embedding](https://mlsanity.com/paper/1803.10963/) (2018)
- [Speaker Recognition Based on Deep Learning: An Overview](https://mlsanity.com/paper/2012.00931/) (2020)
- [ASVspoof 2019: spoofing countermeasures for the detection of synthesized, converted and replayed speech](https://mlsanity.com/paper/2102.05889/) (2021)
- [The BOSARIS Toolkit: Theory, Algorithms and Code for Surviving the New DCF](https://mlsanity.com/paper/1304.2865/) (2013)
- [Learnable MFCCs for Speaker Verification](https://mlsanity.com/paper/2102.10322/) (2021)
- [A Comparative Re-Assessment of Feature Extractors for Deep Speaker Embeddings](https://mlsanity.com/paper/2007.15283/) (2020)

## Abstract (arXiv)

Multi-taper estimators provide low-variance power spectrum estimates that can be used in place of the windowed discrete Fourier transform (DFT) to extract speech features such as mel-frequency cepstral coefficients (MFCCs). Even if past work has reported promising automatic speaker verification (ASV) results with Gaussian mixture model-based classifiers, the performance of multi-taper MFCCs with deep ASV systems remains an open question. Instead of a static-taper design, we propose to optimize the multi-taper estimator jointly with a deep neural network trained for ASV tasks. With a maximum improvement on the SITW corpus of 25.8% in terms of equal error rate over the static-taper, our method helps preserve a balanced level of leakage and variance, providing more robustness.