In issues like hearing impairment, speech therapy and hearing aids play a major role in reducing the impairment. Removal of noise signals from speech signals is a key task in hearing aids as well as in speech therapy. During the transmission of speech signals, several noise components contaminate the actual speech components. This paper addresses a new adaptive speech enhancement (ASE) method based on a modified version of singular spectrum analysis (MSSA). The MSSA generates a reference signal for ASE and makes the ASE is free from feeding reference component. The MSSA adopts three key steps for generating the reference from the contaminated speech only. These are decomposition, grouping and reconstruction. The generated reference is taken as a reference for variable size adaptive learning algorithms. In this work two categories of adaptive learning algorithms are used. They are step variable adaptive learning (SVAL) algorithm and time variable step size adaptive learning (TVAL). Further, sign regressor function is applied to adaptive learning algorithms to reduce the computational complexity of the proposed adaptive learning algorithms. The performance measures of the proposed schemes are calculated in terms of signal to noise ratio improvement (SNRI), excess mean square error (EMSE) and misadjustment (MSD). For cockpit noise these measures are found to be 29.2850, -27.6060 and 0.0758 dB respectively during the experiments using SVAL algorithm. By considering the reduced number of multiplications the sign regressor version of SVAL based ASE method is found to better then the counter parts.

Removing of noise component is an important task in all practical applications like hearing aids, speech therapy etc. In speech communication applications the speech components are contaminated with various types of noises. Separation of speech and noise component is a key issue in hearing aids, speech therapy applications. This paper demonstrates a hybrid version of singular spectrum analysis (SSA) and independent component analysis (ICA) based adaptive noise canceller (ANC) to separate noise and speech components. As ICA is not suitable for single channel sources, SSA is used to map signal channel data to multivariant data. Therefore, SSA based ICA decomposition is used to generate reference for noise cancellation process. Variable Step based adaptive learning algorithm is used to separate noise contaminations from speech signals. To reduce computational complexity of system, sign regressor operation is applied to the data vector of the proposed adaptive learning methodology. Performance measures such as Signal to noise ratio improvement, excess mean square error and misadjustment are calculated for various considered ANCs, their values for crane noise are 29.6633 dB, – 27.4854 dB and 0.2058 respectively. Among the various adaptive learning algorithms, sign regressor based step variable method performs better than the other algorithms. Hence this learning methodology is well suited for hearing aids and speech therapy applications due to its robustness, less computational complexity and filtering ability.

Removal of noise components of speech signals in mobile applications is an important step to facilitate high resolution signals to the user. Throughout the communication method the speech signals are tainted by numerous non stationary noises. The Least Mean Square (LMS) technique is a fundamental adaptive technique usedbroadly in numerouspurposes as anoutcome of its plainness as well as toughness. In LMS technique, an importantfactor is the step size. It bewell-known that if the union rate of the LMS technique will be rapidif the step size is speedy, but the steady-state mean square error (MSE) will raise. On the rival, for the diminutive step size, the steady state MSE will be minute, but the union rate will be conscious. Thus, the step size offers anexchange among the convergence rate and the steady-state MSE of the LMS technique. Build the step size variable before fixed to recover the act of the LMS technique, explicitly, prefer large step size values at the time of the earlyunion of the LMS technique, and usetiny step size values when the structure is near up to its steady state, which results in Normalized LMS (NLMS) algorithms. In this practice the step size is not stable and changes along with the fault signal at that time. The Less mathematical difficulty of the adaptive filter is extremely attractive in speech enhancement purposes. This drop usually accessible by extract either the input data or evaluation fault. The algorithms depend on an extract of fault or data are Sign Regressor (SR) Algorithms. We merge these sign version to various adaptive noise cancellers. SR Weight NLMS (SRWN-LMS), SR Error NLMS (SRENLMS), SR Unbiased LMS (SRUBLMS) algorithms are individual introduced as a quality factor. These Adaptive noise cancellers are compared with esteem to Signal to Noise Ratio Improvement (SNRI).

Noise cancellation from the speech signal is the most importanttask in applications like communications, hearing aids, speech therapy and many others. This allows providing good resolution speech signal to the user. The speech signals are mostlycontaminated due to the several natural as well as manmade noises. As the characteristics of these noises random in its nature filtering techniques with fixed coefficients are not suitable for noise cancellation task. Hence, in this work an adaptive noise canceller algorithmhas driven for enhancement of speech signal applications which has the capability to update its weight coefficients based on the statistical nature of the undesired component in the actual speech signal. In our experiments in order to achieve better convergence rate as well as filtering capability we propose Step Variable Least Mean Square (SVLMS) algorithm instead of constant step parameter. The computational complexity of the speech enhancement process is also a key aspect due to the excessive length of the speech signals in realistic scenario. Hence, to reduce the computational complexity of the proposed mechanism we used Sign Regressor SVLSM (SRSVLMS), which is a hybrid realization of familiar sign regressor algorithm and the proposed SVLMS. Using these two techniques noise cancellation models are developed and tested on real speech signals with unwanted noise contaminations. The experimental outputsconfirm that the SRSVLMS based speech signal enhancement unit performs better than its counterpart with respect to convergence rate, computational complexity and signal to noise ratio increment

No abstract