I got an Azure Kinect DK unit today and I am trying to save recordings for the analysis. I need both audio and video for the analysis. I am able to save the recordings but there is no voice in the recordings. Surprisingly, this page of the Kinect also mentions that it cannot record audio. I wonder if this has a microphone array then how come it does not store audio! Is there any other way to store/log the audio with the video steam?
Asked
Active
Viewed 1,015 times
1
-
So, what how did you go about this? – Bendemann Feb 22 '21 at 21:17
2 Answers
1
The microphone for Azure Kinect DK can be accessed through the Azure Speech Devices SDK
Alberto Vega
- 532
- 4
- 13
0
You can try pyaudio(https://people.csail.mit.edu/hubert/pyaudio/docs/). You can find the corresponding microphone device using get_device_count() and get_device_info_by_index(). Number of channels can also be specified.
It works with my Kinect DK. I changed the number of input channels to 7 since DK has an array of 7 microphones. The result is a nd-array with dimension (X,7), where X is the length of your audio. I am not certain whether each channel corresponds to a unique microphone, but there are variations among the 7 columns of recordings.
FrankBu0616
- 1
- 2