VFA-2026-0015
A queue-ordering race in assembling the agent's own spoken output into conversation history drops the turn's final word from the stored record, even on a normal turn with no interruption, though the caller still hears the full sentence.
What this is
Filed as TTS word drop context race, the name every report of this mode is collected under.
A failure mode is one named thing that goes wrong on a live call. This record carries every public report of it the atlas has found, in each reporter's own words, and a status that moves only when a re-check stores the artifact behind the move. Status fixed: an upstream thread for this mode has been closed as completed.
Evidence
Every independent public report of this mode, newest first: who wrote it, where, when, what they said, and the link back to the post it came from.
...word of a sentence is intermittently dropped from the assistant LLMContext on completely normal, non-interrupted turns. The audio is spoken in full — only the stored transcript is short. Because...
https://github.com/pipecat-ai/pipecat/issues/5119
first seen 2026-07-24 · fetched 2026-07-25T17:50:31Z · body sha256 4891230123214ad1
Each quote is at most 30 words around the reported failure and links to the post it came from. If you wrote one of these and want it off this page, the removal path takes it down within 24 hours.
Hit this one yourself? Add your call to this entry: the stack, the versions, and what the caller heard. It lands as another independent report on this page, with your handle and a link back to what you wrote.
Report another occurrenceThe form is a GitHub issue on the hotato repo. Prefer email? [email protected] reaches a person either way.
What is measurable
Evidence tier: Atlas record: every field on this page traces to a fetched public artifact.
What 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.
- 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.
# the timing incidents in one recording, ranked hotato autopsy ./call.wav
The tier above is what this record establishes on its own: linked sources, an artifact behind every status move, and timestamps that trace to a fetched response. To put a number on your own call, hotato autopsy reads a recording you already have and ranks the timing incidents in it. hotato is MIT and runs locally. getting started.
Status history
A daily re-check reads every source thread on this page. The status moves when that re-check stores the link, the quote and the date behind the move, so every state on this record is one you can open.
- openfixed2026-07-26
issue closed (completed)
https://github.com/pipecat-ai/pipecat/issues/5119
evidence dated 2026-07-25
Discussion
Reader discussion of this mode: what fixed it, what did not, and what the record here is missing. It is a conversation, not part of the evidence above. The reports on this page are attributed and linked to their sources, and a comment is a comment.
Open this discussion on GitHubThe record
References
Every link this entry stands on: the post behind each report, and the artifact behind each status move.
- reporthttps://github.com/pipecat-ai/pipecat/issues/5119 · Sarita-Joshi · GitHub issues · 2026-07-24
Machine surfaces
Removal
If you wrote something quoted on this page and want it gone, one email 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.