Published Paper Details:

The speech signal conduct information about the identity of the speaker. The area of speaker information cares with extract the identity of the person express the utterance. The utility of automatically identifying a speaker is predicated solely on vocal characteristic. This emphasizes on text dependent talker identification, which deals with detecting a specific speaker from a known region. System identifies the user by comparing the code book of speech utterance with those of the stored within the database and lists, which contain the foremost likely speakers, could have given that speech utterance. The speech signal is recorded for number of speakers further the features are extracted. Feature extraction is completed by means of LPC coefficients, calculating AMDF, and DFT. The neural network is train by applying these features as input parameters. The train network corresponds to the output, the input is that the extracted features of the speaker to be identified. The network does the weight adapting and the best match is found to identify the speaker of the region.