Difference between revisions of "Audio"

Latest revision as of 15:12, 20 March 2024

When using audio in your experiment, especially when presenting time-critical stimuli, special care should be taken to optimize the audio settings on multiple levels (hardware, OS, script), as many things can go wrong along the way.

This page outlines some best practices, however we advise to always consult a TSG member if you plan to run an audio experiment in the labs.

Recording

When recording audio for stimuli material or as input for your experiment, please:

Use a high quality microphone, with a polar pattern suitable for your application.
Use a high quality recorder or audio interface, capable of recording at 24bit and 48kHz or higher.
Place the microphone at an appropriate distance from your subject. Set the levels so the audio does not clip (exceeding maximum volume).
Record in a quiet environment.

You can use our Sound Recording Labs for high quality voice recording.

Editing

We recommend using Audacity for editing and converting audio files. Audacity is open-source and fairly easy to use, available here: https://www.audacityteam.org/

Export Settings

We recommend using the following export settings:

File format: .wav (PCM).
Sample Frequency: 44.1kHz.
Bit depth: 16 bit.

The Lab Computer audio output is also set to 16 bit, 44.1kHz. We found that this is good enough for most applications; higher settings will increase file size with limited perceivable quality gains.

When using multiple audio files in your experiment, make sure they all use the same settings for consistent playback in your experiment.

In Audacity, you can set up Macros to automate processing and exporting your audio files: https://manual.audacityteam.org/man/macros.html

Windows Settings

Windows 10 has a habit of automatically enabling audio enhancements when connecting new speakers or headphones. These "enhancements" can distort your audio and cause timing issues. Therefore, please make sure they are turned off:

Right click sound icon on taskbar (next to clock) -> Sounds
Goto Playback tab. Select your audio output device and click "Properties"
Goto Enhancements tab. Make sure "Disable all enhancements" is checked.
Click Apply.

Playback

Psychopy

This is an example of a Python script that plays a .wav file with high time accuracy.

 1from psychopy import sound, core
 2from psychopy import prefs
 3prefs.hardware['audioLib'] = ['PTB']
 4
 5# Path to audio file
 6audio_file = "voice.wav"
 7
 8# Load audio
 9# preBuffer – integer to control streaming/buffering -1 means store all 
10audio = sound.Sound(audio_file,preBuffer=-1)
11
12# Play audio
13audio.play()
14
15# Wait for audio to finish playing
16core.wait(audio.getDuration())
17
18# Close audio
19audio.stop()
20audio.close()

@@ Line 1: / Line 1: @@
-Gebruik van audio in de labs:
+When using audio in your experiment, especially when presenting time-critical stimuli, special care should be taken to optimize the audio settings on multiple levels (hardware, OS, script), as many things can go wrong along the way.
-Bij het gebruik van audio bestanden is het belangrijk om ervoor te zorgen dat de bronnen gelijke instellingen hebben, zodat het afspelen consistent en van hoge kwaliteit is.
+This page outlines some best practices, however we advise to always consult a TSG member if you plan to run an audio experiment in the labs.
-Advies voor het instellen van gelijke instellingen voor audio bestanden:
+==Recording==
+When recording audio for stimuli material or as input for your experiment, please:
+* Use a high quality microphone, with a [https://www.audio-technica.com/en-us/support/a-brief-guide-to-microphones-whats-the-pattern/ polar pattern] suitable for your application.
+* Use a high quality recorder or audio interface, capable of recording at 24bit and 48kHz or higher.
+* Place the microphone at an appropriate distance from your subject. Set the levels so the audio does not clip (exceeding maximum volume).
+* Record in a quiet environment.
-. Bestandsindeling: Het wordt aanbevolen om gebruik te maken van het .wav formaat, omdat het een ongecomprimeerd formaat is en geen verlies van audiokwaliteit veroorzaakt.
+You can use our [[Sound Recording Lab]]s for high quality voice recording.
-. Bemonsteringsfrequentie: Zorg ervoor dat alle audio bestanden dezelfde bemonsteringsfrequentie hebben. Dit verwijst naar het aantal samples per seconde in het audiobestand. Stel de bemonsteringsfrequentie van de bron in op 44.1 kHz, dit is ook de instelling van windows op de labcomputer.
+==Editing==
+We recommend using Audacity for editing and converting audio files. Audacity is open-source and fairly easy to use, available here: https://www.audacityteam.org/
-. Bitsnelheid: De bitsnelheid verwijst naar het aantal bits dat per seconde wordt gebruikt om de audiogegevens op te slaan. Een hogere bitsnelheid resulteert in een betere audiokwaliteit, maar resulteert ook in grotere bestandsgroottes. Gebruik een bitsnelheid van 16 bits, dit biedt een goede balans  tussen geluidskwaliteit en bestandsgrootte.
+===Export Settings===
+We recommend using the following export settings:
+* File format: .wav (PCM).
+* Sample Frequency: 44.1kHz.
+* Bit depth: 16 bit.
-Windows-audioverbeteringen:
+The [[Lab Computer]] audio output is also set to 16 bit, 44.1kHz. We found that this is good enough for most applications; higher settings will increase file size with limited perceivable quality gains.
-Om een betere geluidsweergave op Windows te krijgen schakelen we instelling "Enhance audio" uit. Het uitschakelen van audioverbeteringen geeft een nauwkeurigere en onvervalste weergave van het audio bestand. Deze instelling is standaart gedaan op de aangesloten speakers van de labcomputer, als u een extra speaker of koptelefoon aansluit zal deze instelling nog aangepast/uitgezet moeten worden.
+When using multiple audio files in your experiment, make sure they all use the same settings for consistent playback in your experiment.
-Psychopy script voor tijdaccurate weergave van een .wav bestand:
+In Audacity, you can set up Macros to automate processing and exporting your audio files: https://manual.audacityteam.org/man/macros.html
-Hier is een voorbeeld van een Python-script in Psychopy dat een .wav bestand nauwkeurig afspeelt op basis van tijd:
+==Windows Settings==
+Windows 10 has a habit of automatically enabling '''audio enhancements''' when connecting new speakers or headphones. These "enhancements" can distort your audio and cause timing issues. Therefore, please make sure they are turned off:
+# Right click sound icon on taskbar (next to clock) -> Sounds
+# Goto Playback tab. Select your audio output device and click "Properties"
+# Goto Enhancements tab. Make sure "Disable all enhancements" is checked.
+# Click Apply.
+==Playback==
+=== Psychopy ===
+This is an example of a Python script that plays a .wav file with high time accuracy.
+<syntaxhighlight lang="python" line>
 from psychopy import sound, core
+from psychopy import prefs
+prefs.hardware['audioLib'] = ['PTB']
-# Pad naar het audiobestand
+# Path to audio file
-audio_file = "pad_naar_audio_bestand.wav"
+audio_file = "voice.wav"
-# Laad het audiobestand
+# Load audio
-audio = sound.Sound(audio_file)
+# preBuffer – integer to control streaming/buffering -1 means store all
+audio = sound.Sound(audio_file,preBuffer=-1)
-# Speel het audiobestand af
+# Play audio
 audio.play()
-# Wacht tot het audiobestand is afgespeeld
+# Wait for audio to finish playing
 core.wait(audio.getDuration())
-# Sluit het audiobestand
+# Close audio
 audio.stop()
 audio.close()
+</syntaxhighlight>