Omni-sparsity DNN: Fast Sparsity Optimization for On-Device Streaming E2E ASR via Supernet

by   Haichuan Yang, et al.

From wearables to powerful smart devices, modern automatic speech recognition (ASR) models run on a variety of edge devices with different computational budgets. To navigate the Pareto front of model accuracy vs model size, researchers are trapped in a dilemma of optimizing model accuracy by training and fine-tuning models for each individual edge device while keeping the training GPU-hours tractable. In this paper, we propose Omni-sparsity DNN, where a single neural network can be pruned to generate optimized model for a large range of model sizes. We develop training strategies for Omni-sparsity DNN that allows it to find models along the Pareto front of word-error-rate (WER) vs model size while keeping the training GPU-hours to no more than that of training one singular model. We demonstrate the Omni-sparsity DNN with streaming E2E ASR models. Our results show great saving on training time and resources with similar or better accuracy on LibriSpeech compared to individually pruned sparse models: 2


page 1

page 2

page 3

page 4


Learning a Dual-Mode Speech Recognition Model via Self-Pruning

There is growing interest in unifying the streaming and full-context aut...

TODM: Train Once Deploy Many Efficient Supernet-Based RNN-T Compression For On-device ASR Models

Automatic Speech Recognition (ASR) models need to be optimized for speci...

Dynamic Sparsity Neural Networks for Automatic Speech Recognition

In automatic speech recognition (ASR), model pruning is a widely adopted...

Noisy Training Improves E2E ASR for the Edge

Automatic speech recognition (ASR) has become increasingly ubiquitous on...

A baseline model for computationally inexpensive speech recognition for Kazakh using the Coqui STT framework

Mobile devices are transforming the way people interact with computers, ...

Layer Pruning on Demand with Intermediate CTC

Deploying an end-to-end automatic speech recognition (ASR) model on mobi...

weight-importance sparse training in keyword spotting

Large size models are implemented in recently ASR system to deal with co...

Please sign up or login with your details

Forgot password? Click here to reset