Time-frequency Averaging of Noise Compensation Filters for Asr
نویسنده
چکیده
In this paper we show that time-frequency averaging of noise compensation filters, estimated using well-known techniques such as spectral subtraction, Wiener filtering and Ephraim-Malah approach, improves the performance of automatic speech recognition system in non-stationary noisy environments. The experiments were conducted on a multilingual isolated word recognition task in signal to noise ratio ranging from 5dB to 20dB and clean conditions.
منابع مشابه
Application of Single-Frequency Time-Space Filtering Technique for Seismic Ground Roll and Random Noise Attenuation
Time-frequency filtering is an acceptable technique for attenuating noise in 2-D (time-space) and 3-D (time-space-space) reflection seismic data. The common approach for this purpose is transforming each seismic signal from 1-D time domain to a 2-D time-frequency domain and then denoising the signal by a designed filter and finally transforming back the filtered signal to original time domain. ...
متن کاملParametric Estimates of High Frequency Market Microstructure Noise as an Unsystematic Risk
Noise is essential for the existence of a liquid market, and if noise traders are not present in the market, the trade volume will drop severely and an important aspect of the market philosophy will be lost. However, these noise traders bring noise to the market, and the existence of noise in prices indicates a temporary deviation in prices from their fundamental values. In particular, high-fre...
متن کاملA Multichannel Feature Compensation Approach for Robust ASR in Noisy and Reverberant Environments
In this paper we propose a multichannel feature compensation approach for automatic speech recognition in reverberant and noisy environments. The proposed technique propagates the posterior of the clean signal estimated by a multichannel Wiener filter in short-time Fourier transform (STFT) domain into Mel-frequency cepstrum coefficients (MFCC) domain. The multichannel Wiener filter reduces both...
متن کاملLocal averaging and differentiating of spec
Local frequency and time averaging and differentiating operators, using three neighboring points of critical-band timefrequency plane, are used to process the plane prior to its use in TRAP-based ASR. In that way, five alternative TRAP-based ASR systems (the original one and the time/frequency integrated/differentiated ones)are created. We show that the frequency differentiating operator improv...
متن کاملLow latency IIR digital filter design by using metaheuristic optimization algorithms
Filters are particularly important class of LTI systems. Digital filters have great impact on modern signal processing due to their programmability, reusability, and capacity to reduce noise to a satisfactory level. From the past few decades, IIR digital filter design is an important research field. Design of an IIR digital filter with desired specifications leads to a no convex optimization pr...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2005