DeprecationDeskforced-migration deadlines

OpenAI

OpenAI shuts down whisper-1, gpt-4o-transcribe and audio translations on February 26, 2027

99 days left — Feb 26, 2027

What stops working
The speech-to-text models whisper-1, gpt-4o-transcribe, gpt-4o-mini-transcribe and gpt-4o-transcribe-diarize stop answering API requests. The /v1/audio/translations endpoint (upload audio, get English text back) is removed at the same time. Any app that calls these returns errors after that date.
Who is affected
Every developer or company using the OpenAI API for transcription, captioning, call recording, meeting notes or audio translation. whisper-1 is one of the most widely integrated OpenAI audio features.
Official migration path
Move transcription to gpt-live-transcribe or gpt-transcribe; move audio translation to the /v1/realtime/translations endpoint.
What the official path does not cover
The replacements have different request and response shapes (the realtime translation path is a streaming interface, not a one-shot file upload), different output fields for timestamps and speaker labels, and different pricing and rate limits. Re-testing accuracy on the customer's own audio, rewriting batch pipelines for streaming and updating downstream parsers is the paid work.
Source
OpenAI API Deprecations page (notice dated Aug 26, 2026; read Nov 14, 2026) · announced 2026-08-26 · verified 2026-11-14

Need this done before Feb 26, 2027?

Fixed price, fixed scope, fixed date. From $2,500 for a single integration or workflow; you get the working replacement, the code, and a before/after check. Quote within one business day.

Get a fixed-price quote →

One short form (company, email, what runs on this, needed-by). Reply within one business day.

Not ready yet? Get an email when OpenAI deadlines approach · Firms: get listed for OpenAI work.

Spotted a wrong date? Report it — corrected with the source within one business day.

← All deadlines