Speech recognition allows natural communication between the humansand machines. With Industry 4.0 there is a great demand forsystems that perform this task, since human-machine integrationsare increasingly attractive. Currently, there are several tools and resourcesthat perform this activity, with some companies providingtheir audio recognition services through the Application ProgrammingInterface, such as Microsoft, Google, IBM and Wit. On theother hand, there are offline libraries and open source that can alsobe explored like Vosk. Each company has its business rule and itsspecificity, in this sense it is difficult to know which is the most interestingfor each situation. Thus, a comparison was made betweenspeech recognition services in terms of usability, limitation andprecision. In the comparison, speech recognition performance metricswere used in a set of audios, using the programming languagePython.