End-to-end Recurrent Denoising Autoencoder Embeddings for Speaker Identification
This addresses the problem of speaker recognition in real-world noisy conditions for applications like security or communication systems, but it is incremental as it builds on existing autoencoder and data augmentation methods.
The paper tackled speaker identification in noisy and stressed speech by designing an end-to-end recurrent denoising autoencoder that extracts robust speaker embeddings, and it showed that joint optimization outperforms independent optimization and hand-crafted features under these distortions.
Speech 'in-the-wild' is a handicap for speaker recognition systems due to the variability induced by real-life conditions, such as environmental noise and the emotional state of the speaker. Taking advantage of the principles of representation learning, we aim to design a recurrent denoising autoencoder that extracts robust speaker embeddings from noisy spectrograms to perform speaker identification. The end-to-end proposed architecture uses a feedback loop to encode information regarding the speaker into low-dimensional representations extracted by a spectrogram denoising autoencoder. We employ data augmentation techniques by additively corrupting clean speech with real-life environmental noise in a database containing real stressed speech. Our study presents that the joint optimization of both the denoiser and speaker identification modules outperforms independent optimization of both components under stress and noise distortions as well as hand-crafted features.