discrete cosine transforms; feature extraction; speech recognition; MFCC balanced feature extraction; STRF feature; acoustic features; conventional MFCC; conventional Mel frequency cepstral coefficients; discrete cosine transform; logarithmic operation; mammalian auditory system; noise-free environment; noisy speech recognition; physiological model; recognition rate; scale energies; spectral response; spectral-temporal domain; spectral-temporal receptive fields; temporal response; Decision support systems; Mel frequency cepstral coefficient; Physiology; Speech; Speech processing; Mel frequency cepstral coefficients; spectral-temporal receptive fields; speech recognition;
机译:频谱时域接收场和MFCC平衡特征提取,可增强说话人识别能力
机译:噪声条件下使用MFCC,LPCC,PLP,RASTA-PLP和隐马尔可夫模型分类器的常规和混合特征的鲁棒语音识别系统
机译:基于双谱的二维MFCC识别语音噪声信号及其作为特征提取子系统的遗传算法优化
机译:用于噪声语音识别的光谱 - 时间接收领域和MFCC平衡特征提取
机译:具有高效MFCC提取算法和多混合模型的语音识别IC。
机译:使用自然声音获得的非线性听觉神经元的光谱-时间感受野
机译:MSP-MFCC:节能MFCC功能提取方法,具有用于可佩戴式语音识别应用的混合信号处理架构