Audio¶

How audio works in TiXL: clips on the timeline, the routing and mixing graph, effects, audio-reactive visuals, and what ends up in your exports. Since TiXL 4.3, all of this is built from ordinary operators — there is no separate audio configuration beyond a few project settings.

Audio clips on the timeline¶

Drag an audio file (mp3, wav, ogg) from the Asset Library or your file explorer onto the timeline and it becomes an [AudioClip] operator — a clip you can move, trim, split, loop, mute and delete like any other timeline clip. A project can have many audio clips playing in parallel, across as many layers as you need.

AutoPlay (on by default) makes the clip play whenever the playhead is inside it — no wiring required.
Trimming the clip's edges plays only that part of the file; dragging the start before the file's beginning leaves that stretch silent.
Loop repeats the trimmed source window for the length of the clip. Repeat boundaries show as thin lines in the clip body.
Mute silences the clip; muted clips render faded in the timeline.
Style picks the image drawn in the clip body: a frequency Spectrum, a peak Waveform, or a smoothed Volume Level.
Renaming the operator shows the new name in quotes on the clip; the file name stays visible in the tooltip.

Splitting a clip (Shift+X at the playhead) keeps both halves playing seamlessly, and reconnects the new clip to whatever the original was wired into.

The main soundtrack¶

One clip can be designated the main soundtrack. It renders as the full-width image behind the timeline, drives audio-reactive operators, and defines the duration of exported executables.

To designate it, right-click an audio clip and choose Set as Main Soundtrack, or use Create Soundtrack in Project Settings → Audio. Technically the designation is just the clip's Display parameter set to Background Image — so it travels with the operator, not with any hidden project state.

The main soundtrack always spans its full source content, and its clip block is hidden from the timeline layers: the background image is its representation. To reposition it on the timeline or turn it back into a regular clip, set the operator's Display parameter back to Clip — Project Settings → Audio → Select and focus Main Soundtrack jumps you to the operator.

Routing and mixing¶

By default, clips play on their own and you never see a wire. For mixing, grouping and effects, use the audio graph:

Every [AudioClip] has an AudioReference output. Wire it into an [AudioBus] and the bus takes over the clip's playback level and mixing. The bus is the sink of the graph — it owns a master Volume and reports a Level you can meter.
[CombineAudio] groups several sources under a shared group volume before they reach the bus. Groups can nest.
Effects — [AudioReverb], [AudioEcho], [AudioCompressor] — are inserts: everything flowing through them is processed. Chain them by wiring one into the next.
AudioToneGenerator produces test tones and synth drones directly in the graph; [PlayAudioSample] triggers one-shot samples.

With AutoCollectClips enabled on a bus or a group, unwired sibling audio clips join it implicitly — useful when you want a whole composition's clips to share a volume, a meter, or an effect without wiring each one. Use only one auto-collecting operator per composition.

The bus must be evaluated to play — wire it into your render chain or pin it to an output view.

Video audio¶

Videos play their sound: both PlayVideo and a VideoClip on the timeline decode their file's audio track and play it in sync with the picture. The Volume parameter scales it, and at 0 the audio track isn't decoded at all — a video used purely as a texture costs nothing extra. Like every audio source, both operators have an AudioReference output: wire it into an [AudioBus] or [CombineAudio] to give video sound a group volume, effects or ducking; left unwired it simply plays.

Sound follows the picture:

A PlayVideo that nothing renders is silent; a VideoClip is audible only inside its cut, and trimming or slipping the clip moves its sound along with the frames.
Audio plays during normal forward playback. Scrubbing, reverse and other playback speeds are silent — pitch-corrected audio for those is planned.
Preview proxies carry no audio track, so sound always comes from the original file, proxy preview or not.
Video audio is not yet included in rendered video files or executables (see Export).

Metering and ducking¶

An [AudioLevel] tap measures the signal at the point where it sits in the chain — wire it inline (source → tap → bus), not as a side branch, so it can meter timeline clips. Its level output drives anything: lamp brightness, scale, or a [DuckAudioLevel], which lowers one signal while another is loud — the classic voice-over-over-music setup:

Route the music through a [DuckAudioLevel].
Tap the voice-over group with an [AudioLevel] and wire its level into the duck's control input.
Adjust threshold and amount until the music breathes with the voice.

Audio-reactive visuals¶

AudioReaction reacts to the project's audio analysis — beats, bass, highs — and outputs values to drive parameters. What it listens to depends on the project setup (Project Settings → Timing):

Animation projects analyze the main soundtrack, frame-accurately — also while rendering videos, so exported reactivity matches playback.
Live / Interactive projects analyze the audio input device configured in Project Settings → Audio, with Gain to adjust varying input levels and Decay to shape how quickly reactions fall off.

To react to one source instead of the whole mix, insert the AudioReaction into the graph: wire the source's AudioReference into its Source input and its Result onward into the bus. Wired inline like this it analyzes only what flows through it — for example a video's sound while music plays alongside. As a dead-end side branch it can't analyze timeline clips or video audio; the operator warns when it's wired that way.

Export¶

What you hear is what you get: timeline clips and everything routed through the audio graph — including group volumes, effects and ducking — render into exported videos and executables. The main soundtrack defines an executable's play duration.

The exception for now is video audio: it plays live in the editor but is not yet part of the deterministic export mixdown, so rendered files and executables don't carry it.