Synthesia and sync-3: a catalogue against a count
Synthesia lists each voice with a language name, a code, a gender and an id. sync-3 says ninety-five or more languages and describes no voice. The smaller answer is the one a production can act on. As of 2026-09-22.
| Field | Synthesia | sync-3 | Where they part |
|---|---|---|---|
| Audio source | From the script, or uploaded | Supplied, or read from text | Different answers |
| Voice source | A catalogue voice, or a cloned one | Nothing documented as an input | Different answers |
| Per-character binding | Not documented by the vendor | Not documented by the vendor | Same answer |
| Languages | A catalogue with codes | A count, ninety-five or more | Different answers |
| Lip-sync | The spoken content, framed close | The audio it is given | Different answers |
Inclusion rule. Two entries are given a page together when at least one column puts them at opposite grades of answer. Pairs that agree on every column, or that are both blank throughout, do not get a page. Order. Fixed field order, identical on every side-by-side page.
1A larger number and a smaller answer
Ninety-five or more is far bigger than any catalogue here and settles nothing, because it describes alignment holding across phonetics rather than voices being available. A catalogue with language codes settles a market question in a minute.
This is the clearest case in the register for grading a list above a total. The two numbers are not measuring the same quantity, and the one that looks weaker is the one that answers the question a commissioning meeting asks.
2The difference follows from where the voice comes from
A platform that supplies voices has to enumerate them, because a user has to pick one. A platform handed a waveform has nothing to enumerate; the voice arrived with the recording, so language coverage becomes a claim about sounds rather than about casting.
Neither vendor is being evasive. The shape of the answer follows the shape of the product, and a reader comparing the two numbers without noticing that will reach the wrong conclusion twice.
3Both name a driver for the mouth, and by different means
One generates lip movement from the spoken content and attaches a framing condition to it. The other matches movement to the audio it is given. Two of only five entries in this register that name a driver at all, arrived at from opposite directions.
The framing condition is the rarer sentence. Saying performance improves with the avatar closer to camera names a weak case, and naming a weak case constrains a shot list in a way a capability claim never does.
4Each of them on its own
The column this pair was chosen for is languages, and each entry has a page of its own on it. The full row for either, all five columns with the wording behind each cell, is on its model note.
- Synthesia on languages — a catalogue with codes.
- sync-3 on languages — a count, ninety-five or more.
- Synthesia, all five fields — read from the voices reference and avatar documentation.
- sync-3, all five fields — read from the model documentation.
- LanguagesEach voice is listed with its formal and native language name, a language code, a gender, a name and a voice idpublished as a catalogue with codes
- Voice sourceVoice cloning carries a language list of its own, running from Afrikaans to Zulu and described as the same list used for standalone cloninga cloned voice with its own reach
- A framing condition attached to lip-syncLip sync performs best when the avatar is framed closer in the scene rather than positioned far from the camera
- Languagessync-3 supports 95 or more languages, described as the same coverage as the models before itpublished as a count
- Audio sourceThe accepted pairs are video with audio, video with text, image with audio and image with textsupplied or read from text
5Sources
Every cell above is read from the documentation each vendor publishes, on the dates carried by the two model notes. The whole register on one column is on the field note; entries grouped by the shape of their answer are on the routes. Other pairs: SceneMixer and Kling AI, Runway and MiniMax.