Quicklogic Forums

Has anyone come across example firmware that can be used to stream audio data from the quickfeather board ?

UPDATE: I came across this directory in github that includes files specifically for the audio sensor. It includes an updated dcl_import.ssf with the XML for the audio sensor added. There are also links for the .cpp and .h files for the audio sensor.

It looks like the audio-related code was just committed 2 days ago, so I plan to start by downloading the latest repository and going from there.

https://github.com/QuickLogic-Corp/qorc ... ssi_ai_app


*******************
I'm also looking for this. Nothing found yet.

I requested a webinar or sample project on the QORC/SensiML tutorial today.

You will need to rebuild the firmware to collect Audio - qf_apps/qf_ssi_ai_app/inc/Fw_global_config.h lines 73/74:

#define SSI_SENSOR_SELECT_AUDIO (0) // 1 => Select Audio data for live-streaming or recognition modes Change this to a 1
#define SSI_SENSOR_SELECT_SSSS (1) // 1 => Select SSSS sensor data for live-streaming of recognition modes Change this to a 0

From there, you can build and flash. the Quickfeather should then be sending audio configuration, and then audio data, instead of IMU.

Note - you may also want to update to Data Capture Lab 2020.10.4 (releasing today, 2/2/2021), as it adds audio as a natively supported source rather than needing to import with SSF.

I'm able to now stream the audio data, however when using knowledge pack to recognise sound it doesn't work. There seems to be no output on the Uart terminal. Seems to work when the model is directly flashed through bin file.
Any idea on what's going on ?