Text-to-Speech: You can’t set an exact target duration in the TTS API. Timing is controlled indirectly via the speed setting and SSML breaks, but total length isn’t guaranteed. (elevenlabs.io)
Sound Effects: You can set an explicit duration for each generated effect (0.1–30 seconds). (elevenlabs.io)
SSML pauses: <break time="x.xs" /> for natural pauses up to 3s. (help.elevenlabs.io)
Note: TTS is non‑deterministic; slider values don’t guarantee identical results or exact total runtime. (elevenlabs.io)
Precise alignment/workflows:
Studio timeline gives sentence‑level control and trimming for placement against picture or other tracks. (elevenlabs.io)
Voice Changer (speech‑to‑speech) follows your reference performance; useful when you need delivery and timing to match a guide take. (help.elevenlabs.io)
If you need exact timestamps after generation, use Speech‑to‑Text (Scribe) or Forced Alignment to obtain word‑level timings from the audio. (elevenlabs.io)
Sound Effects duration:
You can explicitly set the clip length (up to 30s) when generating SFX; leaving it unset lets the model choose the length. (elevenlabs.io)
If you can share whether this is for TTS dialogue, SFX, or dubbing-to-video, I can point you to the exact endpoint and payload fields. (elevenlabs.io)
crisp olive
brittle sequoia