A system is provided that provides guidance (e.g., audio content) to a user while exercising. This audio content is tailored to the particular user and is aimed at increasing the effectiveness of the exercising. In some cases, the audio content conveyed to the user can be based on one or more sensors providing various metrics (e.g., heart rate, cadence, breathing rate, galvanic skin response, temperature etc.) regarding the real-time performance of the user. In some cases, the guidance to a user may be downloaded dynamically to the device by a specialized server or wireless connection. In addition, the guidance can be dynamically adjusted based on a variety of factors including data obtained from the sensors. Related apparatus, systems, techniques and articles are also described.