# Breaking Down Power Barriers in On-Device Streaming ASR: Insights and Solutions

arXiv: https://arxiv.org/abs/2402.13076 · HTML: https://mlsanity.com/paper/2402.13076/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 33 references that are in the map.
- First posted on arXiv 20 Feb 2024.

## Builds on (most cited first, 18 of 18)

- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Speech Recognition with Deep Recurrent Neural Networks](https://mlsanity.com/paper/1303.5778/) (2013)
- [Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss](https://mlsanity.com/paper/2002.02562/) (2020)
- [Streaming End-to-end Speech Recognition For Mobile Devices](https://mlsanity.com/paper/1811.06621/) (2018)
- [A Comparative Study on Transformer vs RNN in Speech Applications](https://mlsanity.com/paper/1909.06317/) (2019)
- [End-to-End Speech Recognition: A Survey](https://mlsanity.com/paper/2303.03329/) (2023)
- [Emformer: Efficient Memory Transformer Based Acoustic Model For Low Latency Streaming Speech Recognition](https://mlsanity.com/paper/2010.10759/) (2020)
- [Transformer-based Acoustic Modeling for Hybrid Speech Recognition](https://mlsanity.com/paper/1910.09799/) (2019)
- [Streaming automatic speech recognition with the transformer model](https://mlsanity.com/paper/2001.02674/) (2020)
- [Transformer-Transducer: End-to-End Speech Recognition with Self-Attention](https://mlsanity.com/paper/1910.12977/) (2019)
- [A Better and Faster End-to-End Model for Streaming ASR](https://mlsanity.com/paper/2011.10798/) (2020)
- [Tied & Reduced RNN-T Decoder](https://mlsanity.com/paper/2109.07513/) (2021)
- [Optimizing Speech Recognition For The Edge](https://mlsanity.com/paper/1909.12408/) (2019)
- [Omni-sparsity DNN: Fast Sparsity Optimization for On-Device Streaming E2E ASR via Supernet](https://mlsanity.com/paper/2110.08352/) (2021)
- [Memory-efficient Speech Recognition on Smart Devices](https://mlsanity.com/paper/2102.11531/) (2021)
- [ConvRNN-T: Convolutional Augmented Recurrent Neural Network Transducers for Streaming Speech Recognition](https://mlsanity.com/paper/2209.14868/) (2022)
- [Folding Attention: Memory and Power Optimization for On-Device Transformer-based Streaming Speech Recognition](https://mlsanity.com/paper/2309.07988/) (2023)
- [Factorized Blank Thresholding for Improved Runtime Efficiency of Neural Transducers](https://mlsanity.com/paper/2211.00896/) (2022)

## Built on by (most cited first, 6 of 6)

- [Deep Learning Models in Speech Recognition: Measuring GPU Energy Consumption, Impact of Noise and Model Quantization for Edge Deployment](https://mlsanity.com/paper/2405.01004/) (2024)
- [Basis Selection: Low-Rank Decomposition of Pretrained Large Language Models for Target Applications](https://mlsanity.com/paper/2405.15877/) (2024)
- [IMPACT: Importance-Aware Activation Space Reconstruction](https://mlsanity.com/paper/2507.03828/) (2025)
- [A light-weight and efficient punctuation and word casing prediction model for on-device streaming ASR](https://mlsanity.com/paper/2407.13142/) (2024)
- [Decomposed Trust: Privacy, Adversarial Robustness, Ethics, and Fairness in Low-Rank LLMs](https://mlsanity.com/paper/2511.22099/) (2025)
- [CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition](https://mlsanity.com/paper/2610.19134/) (2026)

## Abstract (arXiv)

Power consumption plays a crucial role in on-device streaming speech recognition, significantly influencing the user experience. This study explores how the configuration of weight parameters in speech recognition models affects their overall energy efficiency. We found that the influence of these parameters on power consumption varies depending on factors such as invocation frequency and memory allocation. Leveraging these insights, we propose design principles that enhance on-device speech recognition models by reducing power consumption with minimal impact on accuracy. Our approach, which adjusts model components based on their specific energy sensitivities, achieves up to 47% lower energy usage while preserving comparable model accuracy and improving real-time performance compared to leading methods.