No speaker device found with index None’ when using external microphone

OS: Windows 11
PsychoPy version: 2026.1.3
Standard Standalone Installation? Yes
Do you want it to also run online? No

Hi everyone,

I’m running an experiment in which participants listen to short audio descriptions of words and respond by saying the words aloud.

The stimuli are presented using an Audio component, and responses are recorded using a Microphone component.

The experiment initially ran without any problems when using the laptop’s built-in microphone and speakers. However, when I switched to an external microphone to improve recording quality, the experiment started crashing with the following error:

No speaker device found with index ‘None’.

This happens even though I have selected the appropriate output device (headphones/laptop speakers) in Device Manager and in each Sound component.

Note: the setup worked during an earlier test when headphones were plugged in and I used the laptop’s built-in microphone. However, after unplugging everything and moving to a different location for data collection, I plugged the external microphone and headphones back in, and the error appeared.

I’ve attached a screenshot of the error.

Could this be related to how PsychoPy is detecting or assigning audio input/output devices, or am I missing something in the setup?

I’m also wondering whether this issue may have been addressed in the more recent 2026.2.4 release. I’m happy to update and test, but I’d also like to understand what is causing the problem.

Any advice would be greatly appreciated. Thanks!

We did make some major changes to how speakers are detected in 2026.2 so updating is certainly worth a try; most notably, we now index speakers by their name on the system rather than by a numeric index, which will likely be relevant to this bug.

It’s surprising that it got an index of None though; what does the device look like in Device Manager? Does it have None as the value for index in the table there too?

Thank you so much for your response!
I can’t access the laptop on which I set up the experiment at the moment, so I checked the lastRun file instead. It shows that the headphones have an index of 8.0.

# initialize 'headpho'
deviceManager.addDevice(
    deviceName='headpho',
    index=8.0,
    deviceClass='psychopy.hardware.speaker.SpeakerDevice',
    resample=True,
    latencyClass=1,
)

So I’m confused about where the None in the error message is coming from. I’ll also try updating to 2026.2 when I have access to the laptop and see whether that resolves the issue.

I think I see why the error message is wrong; it reports the speaker’s name rather than index, even though the error is caused by an unrecognised index. We’ve fixed that aspect in 2026.2 (the same error message now correctly reports the index), so if it still doesn’t work upon updating you’ll at least get the correct error.

After updating, it may be worth deleting and recreating the device in Device Manager to be certain you get a name rather than a numeric index.

Hi!

I’ve actually suddenly found the same issue (ish).

My output is reporting "No default speaker specified in Preferences / Hardware, using first speakr found: Speakers (Realtek (R) Audio).

I then installed the new 2026.2.4 version, and now suddenly my headphones are no longer even recording audio (this worked last week). Is there any way I can contact you individually to try to come up with a solution? I am in the middle of data collection currently, and have some follow up trials coming!

My windows laptop has specific headphones speakers set as their default, but PsychoPy can’t seem to detect them and I cant figure out how to make this work!

Hi there,

I’ve updated PsychoPy and recreated the device, but I’m now encountering another issue.

When I select psychtoolbox as the audio library, I consistently receive the error message shown below. I’ve also noticed that no audio clips are being saved to the folder. The audio clips are saved happen when I switch to the sounddevice library, although this isn’t ideal, as (1) I understand that psychtoolbox has better temporal resolution and (2) the audio stimuli don’t play at all…

I would really appreciate your help in resolving this issue, as I’m currently in the middle of data collection and have been running into several problems.

Many thanks for your help!

Linh

Just to say I have sound problems too (in v2026.2.3 standalone installed for all users - windows 11) - with an mvp builder script playing a very short wav file:

  • Device manager scanning fails to find any speakers
  • playing in a loop with sounddevice selected it plays it properly the first time and then plays nothing on subsequent loops
  • playing in loop with psychotoolbox selected it plays a ‘click’ noise (ie not the proper sound) but it does do this on each loop iteration.

I’m just going to roll back as I don’t have time to fault find but please advise when this is sorted,

Thanks, John

Have you setup microphone and speaker devices in Device Manager? This sounds like the problem is the device that’s chosen as the default.

As of 2026.2.0, PsychoPy does now support low-latency audio via sounddevice so this isn’t necessarily true anymore - sounddevice should be fine for low latency stuff now.

This one is strange - so sounddevice does record, but can’t play audio? Is this true regardless of which speaker you choose in your Sound Component?

Hi,

Yes, I have tried setting up both in the Device manager. Just as the individual who recently posted to the forum reported, I have the exact same issue of no audio being recorded. I posted about this as well in more detail:

I am also in the data collection process, and am lost on what to do, as just last week, the experiment worked great! I’m wondering if either of you have found a solution?

Sounddevice records and plays the first audio stimulus in a loop only. The subsequent trials have no sound, unlike when I choose Psychotoolbox - all audio stimuli play seamlessly (but again, no recordings are saved).

With sounddevice, I’ve tried selecting different output devices (e.g., laptop speaker or headset), but the audio only plays through the laptop speaker, even when I select headphones as the output source.

So I need to be able to either (1) record with PTB or (2) play all audio files in a loop with sounddevice.

I really appreciate your time and support so far!

Best,

Linh