Learning Multiscale Features Directly From Waveforms

Abstract

Deep learning has dramatically improved the performance of speech recognition systems through learning hierarchies of features optimized for the task at hand. However, true end-to-end learning, where features are learned directly from waveforms, has only recently reached the performance of hand-tailored representations based on the Fourier transform. In this paper, we detail an approach to use convolutional filters to push past the inherent tradeoff of temporal and frequency resolution that exists for spectral representations. At increased computational cost, we show that increasing temporal resolution via reduced stride and increasing frequency resolution via additional filters delivers significant performance improvements. Further, we find more efficient representations by simultaneously learning at multiple scales, leading to an overall decrease in word error rate on a difficult internal speech test set by 20.7% relative to networks with the same number of parameters trained on spec

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyzhu2016learning

Related papers

• Learning Waveform-based Acoustic Models Using Deep Variational Convolutional Neural Networks (2019)6.77
• End-to-end Speech Recognition From The Raw Waveform (2018)12.81
• Y-vector: Multiscale Waveform Encoder For Speaker Embedding (2020)8.60
• Speech And Speaker Recognition From Raw Waveform With Sincnet (2018)0.00
• Speaker Recognition From Raw Waveform With Sincnet (2018)20.65
• Raw Waveform Encoder With Multi-scale Globally Attentive Locally Recurrent Networks For End-to-end Speech Recognition (2021)0.00
• Sample-level Deep Convolutional Neural Networks For Music Auto-tagging Using Raw Waveforms (2017)0.00
• Multichannel Speech Enhancement By Raw Waveform-mapping Using Fully Convolutional Networks (2019)12.25