Low cost method to reproduce sound with FPGA

This paper presents a detailed description of the preliminary work on assessment of the idea: incorporate sound reproduction on FPGA (field programmable gate array) without needing extra active electronics. The sound is reproduced using only two digital output bits filtered by a proper designed RC network. The sound data is stored on an external serial memory (MMC or EEPROM) and prepared by specific software, RC sound coder. A soda-machine, with user interface speech enabled, was selected as a concrete implementation example where all the control can be performed on a FPGA. The test application uses about 3800-gate corresponding to 1,9% of the Xilinx Spartan-3 XC3S200 FPGA. Besides soda-machines, there are other applications demanding low cost implementations where this approach can be used, for instance the toyspsila industry.