计算机科学
编码器
任务(项目管理)
语音识别
人工智能
基本事实
钥匙(锁)
信号(编程语言)
监督学习
基音检测算法
模式识别(心理学)
语音处理
人工神经网络
操作系统
计算机安全
经济
管理
程序设计语言
作者
Beat Gfeller,Christian Frank,Dominik Roblek,Matt Sharifi,Marco Tagliasacchi,Mihajlo Velimirović
出处
期刊:IEEE/ACM transactions on audio, speech, and language processing
[Institute of Electrical and Electronics Engineers]
日期:2020-01-01
卷期号:28: 1118-1128
被引量:33
标识
DOI:10.1109/taslp.2020.2982285
摘要
We propose a model to estimate the fundamental frequency in monophonic audio, often referred to as pitch estimation. We acknowledge the fact that obtaining ground truth annotations at the required temporal and frequency resolution is a particularly daunting task. Therefore, we propose to adopt a self-supervised learning technique, which is able to estimate pitch without any form of supervision. The key observation is that pitch shift maps to a simple translation when the audio signal is analysed through the lens of the constant-Q transform (CQT). We design a self-supervised task by feeding two shifted slices of the CQT to the same convolutional encoder, and require that the difference in the outputs is proportional to the corresponding difference in pitch. In addition, we introduce a small model head on top of the encoder, which is able to determine the confidence of the pitch estimate, so as to distinguish between voiced and unvoiced audio. Our results show that the proposed method is able to estimate pitch at a level of accuracy comparable to fully supervised models, both on clean and noisy audio samples, although it does not require access to large labeled datasets.
科研通智能强力驱动
Strongly Powered by AbleSci AI