Yes, uploading a few 30‑minute files is fine (minimum 30 minutes total; 2–3 hours is recommended). The docs also note that if you’re uploading multiple hours, it’s better to split them into ~30‑minute samples for easier upload, and that the number of files doesn’t matter—only total runtime. (elevenlabs.io)
About “auto‑splitting”: the documentation does not state that ElevenLabs automatically splits long recordings into smaller training chunks. What is documented is per‑sample processing (e.g., noise removal and speaker separation) after you upload each file. (elevenlabs.io)
Key references:
Professional Voice Cloning guide (recommendations on total runtime, ~30‑minute samples, and per‑sample processing). (elevenlabs.io)
Accepted file lengths: 30–180 minutes for Professional Voice Cloning. (help.elevenlabs.io)
Will the quality of professional voice cloning improve if I split the 2 hours of audio into small batches, or will I get the same result with 30-minute batches?
final torrent
whole karma