Now days the usage of technology become our daily basic needs just like shelter, food, close. In order to have effective and efficient communication we need to have correct spelling checker. Tigrigna language is a Semitic language spoken in northern part of Ethiopia, which has its own writing system using Fidel it possesses 35 primary characters, each representing a consonant and each having 7 variations in form to indicate the vowel which follows the consonant. Spelling is an import aspect of language writing. Poor spelling can interfere with communication between the writer and the reader. The main problem occurred in spell checker are non-word error and real word errors during writing Tigrigna book, magazine, newspapers. This research work tries to auto correct and suggestion words when the input is non-word error whereas it doesn’t detect real word errors. And the main function of automatic spell checker is providing spell error detection and spell error correction. In this study, we had been collected more than1/2 million words and We had extracted word using Text STAT software. In this study we had collected 787 unique words for testing the system applying the rule-based dictionary look up and morphological analysis technique, and we had achieved Recall 89, Precision 87%, Accuracy 80 %and F-measure 88% based on the experiment conducted. In this study we have also conducted using unsupervised Morphological based approach using Morfessor tool, we achieved Recall 99%, Precision 72 %, Accuracy ,73 %and F-measure 84 % based on the experiment conducted. This study showed that rule-based approach out performs well than the unsupervised approach. This study explores word compounding and inflectional word errors. Generally, increasing the dictionary and morphological rule to the knowledge base and increasing corpus size to the unsupervised approach will increase the performance of the proposed methods.