Recording Real Sounds

To record sounds we find many different tools (software), but we are going to learn the use of Audacity, an open source and free software tool that provides most of the options needed to reproduce sound with microcrontrollers.

With microcontrollers it is possible to reproduce fairly high quality sound, however they are limited in terms of memory space. We cannot store too much sound in the space we get offered. Therefore we need to use sound formats of lower quality that will allow us push up to several seconds of sound without the use of external memory chips.

A sound format that can help us very much is the 8bit PCM WAVE (also called Microsoft unsigned 8bit). This sound format is still having a certain degree of quality to reproduce recorded human voice.

The tricky part is to set up Audacity (or any other program) to export the sound as we want it to be.

NOTE: Download Audacity 1.3.5 or higher to get better features for trimming your sounds at the right length

After exporting

After exporting the sound, we need to transform it into a format that can be played by the microcontroller. In our case, we are going to make an array of data that will contain the whole content of the sound. It is not easy to type the whole sequence containing the sound, therefore you should use this program to accelerate the process. If the program closes with an error message that means that the sound was formated wrongly. Also you should not include any tags in the sound file, that would make the system crash.