dev-jam / opensesame-plugin-audio_low_latency Goto Github PK

Hi
since version 10.5 I have huge audio file (more than 1Go) for 3 or 4s of input.
when i reverse to 10.3 i work again in opensesame.

When starting an experiment on mac, the initialization of the recording fail.

Error: OSException
Could not start audio device
Message: Error opening RawInputStream: Internal PortAudio error [PaErrorCode -9986]
This error occurred in the prepare phase of item record_init.

I think it's a opensesame problem : the application don't ask for the microphone authorization via his Info.plist ( or a plugin problem if you can ask for authorization when installing the plugin)

<key>NSMicrophoneUsageDescription</key>
<string>Needed for record audio with Audio low latency plugin</string>

see : https://developer.apple.com/documentation/bundleresources/information_property_list/nsmicrophoneusagedescription

Important
If the appropriate key is not present in your app’s Info.plist file when your app requests authorization or attempts to use a capture device, the system terminates your app. The Xcode debug console displays a message that explains the reason for the crash

Hello.
first thanks for making this plugin.

I'm trying to use the background record tool, and i get some trouble :

1. In audio_low_latency_record_start, it's quite confusing that it ask for a file in the pool (maybe changing the filepool type in yaml to line_edit ? and some ref in code too )
   I understand now that i have to type the name i want and it will try to create it (except i get an Read-only error from line 172, the only solution is to set the full path to the file, not only from the working dir of the experiment which is perfect when running an experiment).

2. When i setup with a filename, i get the file created in the folder i ask, but when i finish the recording i get this error ( with and without the cache to ram option):

File "/Applications/OpenSesame.app/Contents/Resources/share/opensesame_plugins/audio_low_latency_record_start/audio_low_latency_record_start.py", line 278, in record
wav_file.writeframes(b''.join(frames))
TypeError: sequence item 0: expected a bytes-like object, tuple found

In the verbose information, i only get a "Initializing audio recording" and then a "sending stop signal" . may be some more verbose information are needed ?

my working environment :
system : Mac os 12.4 with apple silicon M1
opensesame : OpenSesame 3.3.12 Lentiform Loewenfeld
audio_low_latency : version 2022.05-1

I can join the opensesesame project, and the log from the verbose option in the record init if needed.

have a nice day,

Sébastien