Control System and Speech Recognition of Exhibition Hall Digital Media Based on Computer Technology

Algorithm 1

Deep Convolutional Neural Network Algorithm.

	Input: One noisy speech utterance
	Output: improved speech presence probability estimation
	Initialize the statistics at the initial frame for every frequency
	for every time frame do
	for every frequency do
	Compute the mask estimation using equations (12) and (13)
	Compute the posterior SNR using the noise estimation in equation (15)
	Compute the priori SNR
	Compute the gain function or improved speech presence probability based on CNN input, hidden and output layer
	end for
	end for