#Also, guys, do we know how long is a segment maximum duration? Based from the ElevenLab guide, it s

1 messages · Page 1 of 1 (latest)

crisp thunderBOT
#
AI-generated Answer for matthieu2

Short answer: there is no published, platform‑enforced “maximum segment duration” for Dubbing or Voiceover Studio.

What’s documented

  • Dubbing Studio lets you split/merge clips and regenerate audio per clip, but no per‑segment time cap is specified. (elevenlabs.io)
  • The Dubbing API allows you to create segments by arbitrary start_time and end_time; again, no duration limit is stated for a single segment. (elevenlabs.io)
  • Global limits are on input length/size (e.g., up to 45 min in the UI; up to 2.5 h via API), not per‑segment length. (elevenlabs.io)
  • Text‑to‑speech request limits are expressed as character caps per model (e.g., up to 40k characters ≈ ~40 min for some models), which implies very long single generations are possible via API. (elevenlabs.io)
  • Best‑practice guidance suggests breaking long text into smaller chunks to maintain quality, but no exact duration is mandated. (elevenlabs.io)

About the “Segmenter: Max Segment duration ms: 20000”

  • I can’t find any public documentation that defines this UI field or ties it to a hard runtime limit. The only “20‑second” limit I can verify is the default inactivity timeout for the TTS WebSocket (unrelated to segment duration). (help.elevenlabs.io)

If you need to work with smaller chunks but keep natural continuity, consider Request Stitching to maintain prosody across multiple requests. (elevenlabs.io)

If you can share where you saw the 45–60s note, I can check whether it’s in an older/alternate guide; it’s not in the current public docs I searched. (elevenlabs.io)

Sources:
Dubbing Studio | ElevenLabs Documentation
Create segment | ElevenLabs Documentation
Dubbing | ElevenLabs Documentation
Models | ElevenLabs Documentation
Troubleshooting | ElevenLabs Documentation
How can I keep the WebSocket open? – ElevenLabs
Stitching multiple requests | ElevenLabs Documentation
Dubbing Studio | ElevenLabs Documentation

primal tendon
#

@proven vale

crisp thunderBOT
#
AI-generated Answer for matthieu2

I can’t find any official ElevenLabs documentation that defines a fixed “maximum segment duration” for Dubbing Studio. The public docs describe timeline clips/segments and paragraph-based editing, and they note a per‑paragraph limit of 5,000 characters, but they do not specify any time cap per segment/clip. If you’re seeing “Segmenter: Max segment duration ms: 20000” in the UI, that value isn’t documented publicly, so I can’t verify it. For an authoritative answer, please contact Support. Sources: What is the timeline in Studio; Are there any limitations to the size of a project in Studio; Introducing Dubbing Studio; How can I contact support. (help.elevenlabs.io)

Sources:
What is the timeline in Studio? – ElevenLabs

proven vale
#

Sorry, I forgot to mention, the questions above are related to transcriptions.... and the (" Segmenter: Max Segment duration ms :20000") is the segment property.

crisp thunderBOT
#
AI-generated Answer for mary_03112

Short answer: I can’t find any official ElevenLabs documentation that sets a fixed “maximum segment duration” for Speech-to-Text transcripts or that explains a “Segmenter: Max Segment duration ms” field. The public docs show no hard per‑segment limit for transcription.

What is documented

  • File limits for Speech‑to‑Text: up to 3 GB and 10 hours per file. No per‑segment duration limit is mentioned. (elevenlabs.io)
  • Internal chunking (for speed) is automatic for long files: audio over ~8 minutes is split and transcribed in parallel (up to 4 internal chunks). This is a processing optimization and not a user‑visible “segment duration” limit. (elevenlabs.io)
  • The Transcript Editor lets you split/merge segments and adjust start/end times, but it does not state any maximum segment length. (elevenlabs.io)
  • The Speech‑to‑Text “Create transcript” API lists all request parameters and likewise does not include any setting for max segment duration. (elevenlabs.io)

Recommendation

  • Treat the “Segmenter: Max Segment duration ms: 20000” you’re seeing as an internal/editor control rather than a published limit. If you need the precise behavior of that UI field, please share a screenshot with support or open a ticket; there’s no public doc for it today. In the meantime, you can freely split/merge segments in the Transcript Editor and export as needed. (elevenlabs.io)

Sources:
Speech to Text | ElevenLabs Documentation
Speech to Text | ElevenLabs Documentation
Transcripts (product guide) | ElevenLabs Documentation
Create transcript | ElevenLabs Documentation
Transcripts (product guide) | ElevenLabs Documentation

primal tendon
proven vale
#

Hi Matthieu, oh, yes, you are right, I remember seeing the AI assistance, the one that looks like a...disk, I think.

crisp thunderBOT
proven vale
#

Thanks ~~~