PURPOSE: To request a voice output to a voice output mechanism at high speed by executing unique management and efficient conversion by a user and reducing the amount of data.

CONSTITUTION: In a computer system 1 composed of a voice output mechanism 2, voice data file 3 and user's program 4, a storage part 5 is provided in the voice output mechanism, voice management information 10 such as a voice data name and a voice identifier or the like is added to voice data 11, a voice data managing mechanism 6 is provided to manage the data as voice information 12, a pre-registering mechanism 8 is provided to store the voice identifier and the voice data in the storage part 5 while referring to pre-registered information 7 for selectively defining the voice data stored in the storage part, and a voice output information converting mechanism 9 is provided to convert the voice output information designated with the voice data name by the user's program 4 into the voice identifier and the voice data 11 while referring to the voice information 12 and the pre-registered information 7 and to request the output to the voice output mechanism 2.