crackling/robotic audio stimuli via Opensesame

ttseng

Hello everyone,

I am doing an alternative forced choice identification task in the fMRI scanner.

Some of the audio stimuli (speech sound - syllables) are combined with broadband noise, and the duration is 500ms. The audio files go through Openseame on PC (Windows 10, but not updating frequently to avoid technical issues with other softwares) to speaker (Opto Automatic Noise Cancellation device) and to the headphone in the scanner. Sometimes when running the Opensesame task (back-end: psycho), the stimuli become crackling/robotic sounds, especially when it is with noise. However, the audio files work well when I use laptop (Windows 10) to connect with devices.

Would anyone knows if I should adjust/notice something with PC when running Opensesame with audio files (.wav)?

Thank you very much!

Fab

Hi @ttseng,

I'm no expert on the audio issue, but here are some suggestions. If the crackling noise runs throughout the sound and is not just some sort of clicking when then sound starts and ends, try increasing the audio buffer size in the backend settings. Low values give you low latency, but for large-ish sound files, they can produce cracking noises. Another suggestion (perhaps to use in combination with the first) would be to use .ogg files instead of .wav files. Wav files are uncompressed but they're heavy and contain a lot of information our hearing system doesn't actually detect. The ogg format uses compression but the sound quality is excellent provided the files use a good bit rate. Because ogg files are smaller, they might not saturate the audio buffer as much as wav files (at least that seems logical to me). Try these steps. Hopefully it'll help.

Good luck!

Fabrice.