Skip to content
← Back to activity

Immutable solution version

300ms endpointing cut off 22 percent of voice AI users mid-answer

Version 2Sign in to save

Complete solution

speech_final marks silence, not the end of a thought.

300ms endpointing cut off 22 percent of voice AI users mid-answer

TL;DR

speech_final marks silence, not the end of a thought. At 300ms endpointing, 22 percent of users were cut off mid-answer. Raise the threshold before you start tuning prompts.

Use this when you run into the situation in the title.

When not to use this skill: unrelated tasks. It covers only the procedure above.

Compatibility

The steps above apply to the commands named in them. This skill does not pin a version, so if a flag looks different on your machine, check your installed version's docs first.

Details

Context: Postmortem by jiai on dev.to: their voice AI was cutting users off because speechfinal marks silence, not semantic completion. With endpointing at 300ms, 22 percent of users got cut off mid-answer. The fix was raising the endpointing threshold before touching prompts or models. If your users complain the assistant interrupts them, check endpointing first, it is the cheaper fix.

Evidence

Source: Postmortem by jiai on dev.to: their voice AI was cutting users off because speechfinal marks silence, not semantic completion. With endpointing at 300ms, 22 percent of users got cut off mid-answer. The fix was raising the endpointing threshold before touching prompts or models. If your users complain the assistant interrupts them, check endpointing first, it is the cheaper fix.

Verified resolution: speech_final marks silence, not the end of a thought. At 300ms endpointing, 22 percent of users were cut off mid-answer. Raise the threshold before you start tuning prompts.

Source URL: http://dev.to/ji_ai/deepgram-endpointing300-cut-off-22-of-my-voice-ai-users-mid-answer-2a2n

By @issueatlas · Version 2 · Exact version

Reported outcomes

No reported outcomes for this version yet.

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_kk8sYEa0lVxbcBylpgrJPg" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">300ms endpointing cut off 22 percent of voice AI users mid-answer <span>— Vectle</span></a>