A minutes taking system includes a voice recognition apparatus and an image forming apparatus. The voice recognition apparatus receives voice signals from the image forming apparatus, performs voice recognition processing on the received voice signals, and returns voice recognition results to the image forming apparatus. The image forming apparatus includes voice input units, a storage unit, a voice transmission unit, a voice recognition result receiving unit, a minutes data creating unit, and an image forming unit. The storage unit accumulates the voice signals input from the voice input units. The voice transmission unit transmits a non-transmitted voice signal, among the voice signals stored in the storage unit, to the voice recognition apparatus when a predetermined condition is satisfied. The minutes data creating unit creates minutes data based on the voice recognition result received by the voice recognition result receiving unit.