Automatic speech recognition (ASR) is an automated process that inputs human speech and tries to find out what is said. ASR is useful, for example, in speech-to-text applications (dictation, meeting transcription, etc.), speech-controlled interfaces, search engines for large speech or video archives, and speech-to-speech translation. Punjabi is the 10th most widely spoken language in the world. No considerable work has been done on Punjabi language for automatic speech recognition. In the present work Automatic speech recognition system is developed for isolated words using EEMD and Neural Network in which features are extracted using EEMD and segmentation is done. The Punjabi speech for isolated words is converted to Punjabi text. The aim of the work is to check the accuracy of the EEMD algorithm with noisy signals in contrast of the Speech Recognition. We proceed as detecting the noise level and segmenting the signal for the further processing. Ensemble empirical mode decomposition (EEMD) is a noise assisted method and also a significant improvement on empirical mode decomposition (EMD).