End-to-end Deep Neural Network Age Estimation

Abstract

In this paper, we apply the recently proposed x-vector neural network architecture for the task of age estimation. This architecture maps a variable length utterance into a fixed dimensional embedding which retains the relevant sequence level information. This is achieved by a temporal pooling layer. From the embedding, a series of layers is applied to make predictions. The full network is trained end-to-end in a discriminative fashion. This kind of network is starting to outperform the state-ofthe-art i-vector embeddings in tasks like speaker and language recognition. Motivated by this, we investigated the optimum way to train x-vectors for the age estimation task. Despite that a regression objective is typical for this task, we found that optimizing a mixture of classification and regression losses provides better results. We trained our models on the NIST SRE08 dataset and evaluated on SRE10. The proposed approach improved mean absolute error (MAE) by 12% w.r.t the i-vector baseline.

Age-VOX-Celeb: Multi-Modal Corpus for Facial and Speech Estimation

“…The target number of age bins was 101, ranging from 0 to 100, for both facial and speech age estimation tasks. No label aggregation of age bins was done, unlike a previous work [13]. We adopted the Adam algorithm for optimization with an initial learning rate of 0.001.…”

“…In contrast, relatively small corpora have been used for speech age estimation tasks compared with facial age corpora. For example, NIST Speaker recognition evaluations (SRE) 2008 and 2010, which are standard corpora used for speech age estimation task [11,12,13,14], contain only 1688 speakers. Unfortunately, the recording condition in these corpora is limited to telephone speech.…”

Privacy Implications of Voice and Speech Analysis – Information Disclosure by Inference

“…During puberty, vocal cords are thickened and elongated, the larynx descends, and the vocal tract is lengthened [15]. In adults, agerelated physiological changes continue to systematically transform speech parameters, such as pitch, formant frequencies, speech rate, and sound pressure [28,84].…”

“…Automated approaches have been proposed to predict a target's age range (e.g., child, adolescent, adult, senior) or actual year of birth based on such measures [28,85]. In [85], researchers were able to estimate the age of male and female speakers with a mean absolute error of 4.7 years.…”

Speaker Recognition System based on Age-related Features using Convolutional and Deep Neural Networks

“…• Optimal way of training the x-vectors for the age estimation task is proposed. The training on the NIST SRE08 dataset is performed and testing is done against SRE10 (Ghahremani, et al 2018). The implementation is performed based on Series of Time Delay Layers, a part of the DNN followed by temporal pooling layer that summarized the feature sequence into a single fixed dimension embedding was further fed into the feed-forward layers.…”