Sentioscope

Speech and voice controls, as each vendor documents them

The fields kept for every model note

Five fields are kept for every model here, and each has a page saying what the column records and what it cannot settle. Most entries fill one or two and leave the rest blank, and the blanks are the part worth reading. As of 2026-09-12.

The five fields, in the order a production needs them answeredEach field is a separate published claim, and a model can satisfy one and say nothing about the next. Sound first, then who it sounds like, then whether that survives the next episode, then which language it is delivered in, then what moves the mouth.Settled before dialogue is writtenAudio sourceDoes speech leave the model with the picture, or does the shotarrive silent.Voice sourceCan a production hand over a preset choice or a clip to decidewho it sounds like.Per-character bindingDoes that voice belong to the character, or is it re-establishedon every call.LanguagesWhich languages are named for the spoken performance, ratherthan for the menus.Lip-syncIs anything named as the thing a generated mouth is following.Each field is its own page
Fig. 1 Read downwards, the columns are the questions a dialogue scene puts to a vendor before anybody writes a line of it.

The names in the first column of every model note — audio source, languages, voice source, per-character binding, lip-sync — were chosen because a production has to settle them all before a line of dialogue is written. Each has a page here, because a model can fill one and leave the next untouched, and a column read on its own shows a pattern no single row does.

The order is the order the questions arrive in. Whether sound comes out of the model at all decides whether a dubbing stage exists. Whether a voice can be supplied decides whether a character can be cast. Whether that voice persists decides whether a season holds together. Whether a language is named decides which market is reachable.

Each page carries every entry the speech table carries, with the vendor wording behind each cell and the date it was read. What the field pages add is why the column exists at all and what a filled cell still fails to promise.

1The fields

  • Audio source — Audio source records whether a model returns speech in the same pass as the picture or leaves the shot silent, with the published wording behind each cell.
  • Voice source — Voice source records what a production can supply so speech follows a chosen voice: a preset library, a short reference clip, or nothing published.
  • Per-character binding — Per-character binding records whether a voice belongs to a reusable character or is supplied again on every call, and what that costs across a series.
  • Languages — Languages records which spoken languages a vendor names for dialogue, and whether the documentation gives a list to plan against or only a count.
  • Lip-sync — Lip-sync records whether a vendor names something that drives the mouth, claims the feature without a driver, or leaves the question off the page.

A field is added when enough vendors publish something comparable; where nobody publishes, the gap stays a question in the notes rather than becoming a column of blanks.

Other notes: Speech controls, Models, Routes, Side by side, Questions, Terms, Learn, Data. What counts as a documented control is set out on the reading page.