| Algorithmic Review for Voice and Speech Recognition. |
| Swapnil R. Kurkute And D. G. Agrawal |
| Abstract : |
| In the noisy environment there is presence of background acoustic noise
such as noise produced by vehicle engine inside a vehicle. Also there is
utmost requirement to ensure negligible noise components in the recorded
speech signal. Due to this speech communication or recording is greatly
affected.
One possible way to overcome this problem or satisfy a requirement
to obtain a better recording of the desired signal is the use of a simple noise
canceller. [6]
To fulfill this requirement generally adaptive algorithms are used
which converge rapidly. There are two best choice as adaptive filtering i.e.
recursive least squares (RLS) algorithm & normalized least mean squares
(NLMS) algorithm. Unfortunately this algorithm has high computational
complexity, stability & poor adaptive rate problems respectively.
Proposed algorithm is based on adaptive filtering with averaging
(AFA) used for noise cancellation. The main advantages of AFA algorithm
could be summarized as follows. It has low computational complexity at
the same time high convergence rate comparable to that of the RLS
algorithm and possible robustness in fixed-point implementations. The
algorithm is illustrated on car and office noise added to speech data. [1] |
| Show Full Text : |  | | Facility : |   | |
|
|
|