Robust Wake-Up Word Detection by Two-stage Multi-resolution Ensembles

1 code implementation • 17 Oct 2023 • Fernando López, Jordi Luque, Carlos Segura, Pablo Gómez

It employs two models: a lightweight on-device model for real-time processing of the audio stream and a verification model on the server-side, which is an ensemble of heterogeneous architectures that refine detection.

Iterative pseudo-forced alignment by acoustic CTC loss for self-supervised ASR domain adaptation

1 code implementation • 27 Oct 2022 • Fernando López, Jordi Luque

The alignments are computed iteratively upon a corpus of broadcast TV.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +2

Speech Enhancement for Wake-Up-Word detection in Voice Assistants

no code implementations • 29 Jan 2021 • David Bonet, Guillermo Cámbara, Fernando López, Pablo Gómez, Carlos Segura, Jordi Luque

Keyword spotting and in particular Wake-Up-Word (WUW) detection is a very important task for voice assistants.

Data Augmentation Denoising +2