This article answers a variety of Frequently Asked Questions (FAQs) about working with Sansar audio:
Here are some quick tips:
- Your scene should sound balanced, not unusually loud or soft. To calibrate your relative audio levels:
- Visit various experiences by other creators, including our Sansar Studios experiences.
- Test using voice chat with others, and make sure you can hear each other clearly.
- The default loudness settings for audio emitters and the scene's Background Sound (found in Scene Settings) are intended to be reasonable starting points, and depending on the original sound content, are useful for subtler sounds meant to play in the background. You can increase their loudness as needed for foreground sounds meant to get attention, although we recommend you exercise caution against making sudden sounds too loud, since it may startle visitors and make it difficult to voice chat.
- Set objects to different audio materials so that footstep sounds and reverb are treated appropriately, and if you've got the visuals done and are iterating audio, be sure to enable Compute Scene Reverb (natural echo) in your scene. Otherwise, you won't hear the full effect of audio materials.
- Think of your scene as having "audio zones" that transition smoothly from one to another. Be mindful of the audio mix sounding too cluttered, or music/tonal "key clashes" making visitors feel uneasy (unless that's your intent!). Here are some good articles on the topic that address both the technical and psychological impacts of thoughtful "audio zoning":
- Dense Clarity — Clear Density - Key considerations from Walter Murch.
- Creating the Soundtracks to Visitor Experiences - How music (and other sound) can affect mood.
- How Mr. Q Manufactured Emotion - More lessons from Disney.
- Walk through your experience in VR mode or ask a friend to visit your experience in VR. Turn your head around, pay attention to whether sounds match where they are coming from. Remember, especially in VR mode, audio can add a lot of immersion and value. Even subtle audio (such as "almost not there" room tone) can contribute to "presence" and the feeling of "being there".
For a full list of compatible file formats, please see Uploading items in Sansar.
|Note: Non-standard files in the formats on the list may not work, such as those that are using an additional compression format, or other unorthodox modifications.|
We recommend the following specs, which are common in the audio production and game industries:
- Format: Uncompressed WAV, AIF, or FLAC
- Channels: 1 (mono), 2 (stereo/binaural), or 4 (ambisonic).
- Sample Rate: 48 or 44.1 kHz.
If you're having difficulties uploading a sound file, try an audio converter to convert it into the above specs.
We haven't placed a strict cap on the size of .wav files in Sansar.
If you're on a slower/unreliable Internet connection, an otherwise valid sound may show an "Upload failed" error. We've seen this when uploading longer ambiences or songs that are several minutes in length. In the meantime, retrying may eventually work. Or as a workaround:
- Encode the .wav into an .ogg or .mp3, so it has a smaller file size.
- Try re-uploading. This is not officially supported and may result in lesser sound quality, but trust your ears.
Longer sounds need more memory for playback. This may be an issue if you visit an experience with many long sounds and are on a computer that doesn't meet our recommended system specs.
We are continually tuning Sansar's resource usage, so over time you should see reliability improve. You can do the easy Sourceforge Internet Speed Test to show our Sansar team your connection results.
Sansar has an auto-mixing system that we're continuing to refine, so you can focus on delivering a great sonic experience without worrying about too many manual adjustments. This includes prioritizing some sounds so you can hear them better without the whole mix cluttering up, also known as ducking.
Voice chat and scene-wide Background Sound are not affected by the auto-mixer.
For more information, see the "IMPORTANT:" note at the bottom of The Audio Emitter article.
The Background Sound dropdown lets you choose one sound to loop indefinitely in your experience. Make sure this sound is a seamless loop, otherwise a hiccup or rough gap will be noticeably distracting each time it plays.
This sound is non-spatialized — unlike in-scene emitters — and is fixed to the camera position, meaning it's always heard at a constant volume, and doesn't change relative to your context. Make sure it doesn't drown out other sounds. Background Sound is effectively a non-diegetic sound that does not originate from within the scene. For example, it would be odd to build a train station and use a steam engine as Background Sound, only to feel an uncomfortable disconnect as you walk far away from the train in the scene while the steam engine keeps playing loudly.
We generally recommend limiting its use to those scenarios. In many cases, it is better to use an audio volume, as explained in The Audio Emitter.
With mono (1-channel) sounds, no.
With stereo (2-channel) and ambisonic sounds, the orientation is affected. For example, say you have dragged a stereo sound from inventory into a scene so it appears as an emitter and the sound contains speech on the left channel and music on the right channel, clearly separated (also known as "hard panned"). If you rotate this emitter 180 degrees then publish your scene, then the left/right channels will flip relative to the listener. Experiment and hear for yourself!
The falloff distance is modeled to simulate how sound naturally travels in the physical world.
It cannot be changed, but audio volumes cover a number of situations in which sound needs to be sectioned into specific spaces.
We recommend that you change an audio emitter from a point emitter to a volumetric emitter (a spheroid or cuboid audio volume), then set its dimensions accordingly. This is explained in The Audio Emitter.