VAD fires but STT yields no transcript

fixed

2 independent reportsacross 2 sourceson 5 stackslast reported 2026-08-07

What this is

Voice-activity detection registers that the caller is speaking, but the speech-to-text stage fails to produce any transcript even though the same audio is clearly audible in a separate recording of the call.

Status fixed: an upstream thread for this mode has been closed as completed.

Evidence

Pulkit07292026-08-07deepgrampipecatwhisperGitHub issuesgithub.com/…/5240

...a cascade pipeline, what is the intended way to handle a user turn where VAD detects speech but STT returns no transcript at all? Right now the bot just goes...

First seen 2026-08-07 · our copy fetched 2026-08-07T09:10:25Z · SHA-256 of that copy begins 6ae599f6890eda62… · profile

Ganesh_Krishnan2026-05-14deepgramelevenlabslivekitcommunity.livekit.iocommunity.…sponse/1124

...usable STT transcripts. Example pattern from my logs: User started speaking STT HEALTH CHECK: VAD detected speech 7s ago but no STT transcripts received for christopher! Nudging. Later, some partial...

First seen 2026-05-14 · our copy fetched 2026-07-25T18:25:28Z · SHA-256 of that copy begins 5995abed33177b26… · profile

Quotes run to 30 words around the reported failure and link to the post.Add your call to this entry Take a quote down within 24 hours.

What is measurable

Signal	This record stands on its own linked sources. To put a number on your own call, `hotato autopsy ./call.wav` reads a recording you already have and ranks the timing incidents in it.
Establishes	Every report below links to the original post and names the handle that wrote it; the quote is at most 30 words around the reported failure. What else it establishes, and what your own trace settles The status changes only when a re-check stores the evidence link, the quote and the date behind the change. Every timestamp comes from the fetched artifact it describes, recorded with the sha256 of the response body it was read from. What your trace settles Reproducing the mode on your own stack is the next step: the reports below carry the configuration each reporter was running.
Tier	Atlas record: every field on this page traces to a fetched public artifact. hotato is MIT and runs locally, getting started.

Status history

Date	State	What moved it
2026-08-10	acknowledgedfixed	issue closed (completed) evidence dated 2026-08-10
2026-08-08	fixedacknowledged	topic marked solved evidence dated 2026-05-16
2026-07-26	openfixed	topic marked solved evidence dated 2026-05-16

Mode	Status	Shares
Endpoint misdetection	acknowledged	reported on deepgram; reported on elevenlabs; reported on livekit; reported on pipecat; reported on whisper; reported on github; reported on livekit_forum
Barge-in teardown race	acknowledged	reported on deepgram; reported on elevenlabs; reported on livekit; reported on pipecat; reported on github; reported on livekit_forum
VAD background noise mistaken for caller speech	open	reported on deepgram; reported on elevenlabs; reported on livekit; reported on github

Discussion

Reader discussion of this mode: what fixed it and what did not. The reports above are attributed and linked; a comment is a comment.

Open this discussion on GitHub

The record

Ids, digests, every link this entry stands on, and how to take a quote down

Atlas id	`VFA-2026-0013`
Mode id	`pat_786ae449b5decd6ecd7f7c5a`
Content digest	`sha256:786ae449b5decd6ecd7f7c5a388af7b5e4ef1616815d7ad3f33ac75734ae854e`
First reported	`2026-05-14`
Last reported	`2026-08-07`
First seen by the atlas	`2026-05-14T15:27:21.632Z`
Last re-checked	`2026-08-07T08:33:32Z`
Reported on	deepgramelevenlabslivekitpipecatwhisper
Sources	GitHub issues, community.livekit.io
Scope	2 independent public reports across 2 sources. This entry states where the mode has been reported; where else it can occur is what the next report settles.
Resolved	A stored artifact shows this mode closed upstream: 2026-05-16 2026-08-10. The quote and the date behind each move are in the status history above.
References	reporthttps://github.com/pipecat-ai/pipecat/issues/5240 · Pulkit0729 · GitHub issues · 2026-08-07 reporthttps://community.livekit.io/t/participant-was-audible-in-room-demo-recording-but-livekit-agent-heard-very-low-audio-and-stt-vad-pipeline-missed-most-of-the-response/1124 · Ganesh_Krishnan · community.livekit.io · 2026-05-14
Machine surfaces	feed.json feed.atom feed.schema.json atlas.llms.txt
Removal	If you wrote something quoted on this page and want it gone, one email to [email protected] does it. The removal path takes the quote, the handle and the link off this page, out of the feeds and out of the llms.txt within 24 hours, and tombstones the id so the entry cannot quietly return at a new address.