迈向超低功率的神经形态语音增强与尖端-全子网
概括
这项研究介绍了Spiking-FullSubNet,这是一个超低功率的语音增强系统,使用由大脑启发的尖端神经网络 (SNN). 它为边缘设备实现了卓越的语音质量和能源效率,赢得了英特尔N-DNS挑战赛.
科学领域:
- 人工智能的人工智能
- 信号处理 信号处理
- 神经科学是一个神经科学.
背景情况:
- 深度学习方法显著改善语音增强 (SE),但对于边缘设备来说,它们在计算上昂贵.
- 现有的SE系统在高功耗方面扎,这限制了它们在耳机和助听器等设备中的使用.
- 需要高效的SE解决方案,在资源有限的平台上保持高性能.
研究的目的:
- 提出一种超低功率的语音增强系统,使用尖端神经网络 (SNN).
- 为边缘应用程序开发一种新的SE方法,以平衡性能与计算效率.
- 为了证明大脑启发的计算对实时音频处理的有效性.
主要方法:
- 开发了基于SNN的Spiking-FullSubNet系统,采用全频段和子频段融合策略.
- 引入了一种灵感来自人类听觉系统灵敏度的频率分区方法,以优化子频段建模.
- 整合了一种新的尖端神经元模型,用于动态信息集成和遗忘,增强时间处理.
主要成果:
- 在英特尔神经形态深度噪音抑制 (N-DNS) 挑战数据集上,Spiking-FullSubNet实现了最先进的性能.
- 与现有方法相比,该系统在语音质量和能源效率方面都取得了显著的改善.
- 拟议的系统赢得了英特尔N-DNS挑战赛 (算法轨道),验证了其有效性.
结论:
- Spiking-FullSubNet提供了一个可行的解决方案,用于在边缘的超低功率语音增强.
- 由大脑启发的SNN为开发高效和高性能音频处理系统提供了有希望的方向.
- 公共可用的代码和模型促进了低功率SE的进一步研究和开发.
相关概念视频
01:23Propagation of Action Potentials
4.9K
The propagation of an action potential refers to the process by which a nerve impulse, or "action potential," travels along a neuron.
Neurons (nerve cells) have a resting membrane potential, with a slightly negative charge inside compared to outside. This is maintained by ion channels, such as sodium (Na+) and potassium (K+) channels, which control the flow of ions. When a stimulus, like a touch or a signal from another neuron, triggers the neuron, sodium channels open, allowing sodium ions to...
Neurons (nerve cells) have a resting membrane potential, with a slightly negative charge inside compared to outside. This is maintained by ion channels, such as sodium (Na+) and potassium (K+) channels, which control the flow of ions. When a stimulus, like a touch or a signal from another neuron, triggers the neuron, sodium channels open, allowing sodium ions to...
4.9K
01:26Linear Approximation in Frequency Domain
80
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
80