Best voice cloning TTS in 2026: Sume's API has no cloning endpoint
ElevenLabs v4 reportedly clones from 10 seconds and Gemini lists voice replication. Sume's API reference has no cloning endpoint; TTS takes avatar or voice ids.
If you need to clone a voice through an API, Sume's API is not the tool. Sume's public API reference documents no voice-cloning endpoint; the TTS API picks a voice by avatar_id, avatar_handle or voice.id. TechCrunch reported that ElevenLabs v4 clones a voice from 10 seconds of audio, and the Gemini changelog lists voice replication for Gemini 3.8 TTS, so those are the vendors to read for cloning.
No page here ranks them, because quality is for your ears to judge and consent rules vary.
What do the reports say about cloning?
Only dated, stated facts are in the table.
| Vendor | Reported claim | Source |
|---|---|---|
| ElevenLabs v4 | Clone from 10 s of audio | TechCrunch |
| Gemini 3.8 TTS | Voice replication listed; unavailable in IL, TX, EEA, UK, CH, IN | Changelog, Digital Applied |
| Sume API | No cloning endpoint in the API reference; voice ids only | Sume API reference |
What should I ask before cloning anyone?
Ask for written consent from the speaker for the specific use, keep it with the files, and avoid cloning anyone who has not agreed. Check the vendor's terms and your local law; this page is not legal advice.
What does Sume do for voice instead?
Use a voice id for consistent narration across videos, and read the job record to confirm which voice ran. If you must clone a voice through an API, use a vendor whose API offers it, and keep the consent record with the files.
Sources
Related posts
More in Comparisons
- Captions app background stroke and templates vs Sume design
Captions app added a background stroke and cloud caption templates in June 2026. Sume's per-request design object covers stroke and card; saved looks are yours.
- Cartesia acceptable use: the explicit consent clause for clones
Cartesia's policy allows only your own voice or others' with explicit consent. What that means for a voice you clone on Sume, which runs on Sonic.
- Cartesia Ink STT hours per plan vs Sume STT $0.60 per hour
Cartesia plans include about 9 to 741 hours of Ink speech-to-text. That works out near $0.40 to $0.54 an hour; Sume STT is $0.60 an hour. Concurrency compared.
- Chatterbox Multilingual V3 voice cloning (MIT) vs Sume voice ids
Chatterbox is MIT-licensed TTS with 10-second voice cloning and Perth watermarking. How it differs from Sume, which takes a voice id and no clone upload.
Written by Sume