What LTX Studio documents about speech

LTX Studio (ltx.io) documents joint audio and video generation on LTX-2.5 and an audio-to-video route in the other direction, and attaches voices to character Elements so a character keeps its voice between shots. As of 2026-09-12.

Inclusion rule. Fields are the same five for every model here; a field the vendor does not document is recorded as not documented rather than inferred from generated output. Order. Fixed field order, identical on every model page.

1A documented route in both directions, which the others do not describe

Every other model here generates sound alongside the picture and stops there. This one also documents audio-to-video, where an existing track drives the generation rather than being produced by it. For advertising and music work that inverts the usual order of operations: the audio is the fixed element and the picture is fitted to it.

What it means for drama is less obvious and the documentation does not say. A dialogue track recorded by an actor is an existing track, and whether the route is intended to accept one is not addressed anywhere public. It is recorded here as a documented capability rather than as a workflow the vendor endorses.

2Voices live on the character, which is the durable form

Attaching a voice to a character Element puts it in the same place as the character's appearance, so a production defines the cast once and draws on it. That is the arrangement that survives a long shoot, because nothing has to be re-specified per shot and there is no per-call step to forget.

Three fields stay empty against it. No language list, no statement about where the voices themselves come from, and nothing on lip-sync. A production can therefore expect a character to sound like itself, without being able to say in advance what it will sound like or in which languages it can speak.

3This entry, one column at a time

Each of these stays inside a single field: what LTX Studio puts there, what the wording settles, and what it leaves for a take to answer.

• Audio source — with the picture, and audio to video.
• Voice source — nothing documented as an input.
• Per-character binding — on the character element.

4Read against another entry

Each of these puts LTX Studio beside one other entry on a column where the two land at opposite grades of answer.

• HeyGen and LTX Studio — on per-character binding.
• LTX Studio and Vidu — on per-character binding.

5Where this model sits against the rest

Inclusion rule. Models whose vendor documents something about generated speech. A model with nothing published on the point is still listed, with the cell marked hollow. Order. Alphabetical by model name.

Inclusion rule. Models whose vendor documents something about generated speech. A model with nothing published on the point is still listed, with the cell marked hollow. Order. Alphabetical by model name.

Inclusion rule. Models whose vendor documents something about generated speech. A model with nothing published on the point is still listed, with the cell marked hollow. Order. Alphabetical by model name.

6Sources

Taken from the product page at ltx.io on 2026-09-12. Every entry against the same five fields sits on the speech controls page; the rule behind each field is on how read. The entries either side of this one: Luma Ray, MiniMax.

• Audio source
  Joint audio and video generation, plus audio-to-video, on LTX-2.5two directionsLTX Studio / recorded 2026-09-12
• Per-character binding
  Voices are attached to character Elementstied to the element systemLTX Studio / recorded 2026-09-12