Dev.to AI πŸ€– Ai πŸ‘ 0 πŸ“– 1 min read

OpenAI shuts down whisper-1 and its other transcription models on 26 February 2027: a migration checklist

If your app transcribes audio with OpenAI, check which model it calls. OpenAI's speech-to-text guide says several models are deprecated and will shut down on 26 February 2027, and whisper-1 is only one of them. Written

If your app transcribes audio with OpenAI, check which model it calls. OpenAI's speech-to-text guide says several models are deprecated and will shut down on 26 February 2027, and whisper-1 is only one of them.

Written with AI assistance. Read from OpenAI's documentation on 18 November 2026; confirm there before you plan, since schedules can move.

What the guide says

The deprecated list on that page:

  • whisper-1
  • gpt-4o-transcribe
  • gpt-4o-mini-transcribe
  • gpt-4o-transcribe-diarize
  • the /v1/audio/translations endpoint

All are listed as shutting down on 26 February 2027. The guide's recommended starting point for transcribing recorded speech is gpt-transcribe. The same page also notes a 25 MB file limit and that its examples for speaker labels, word timestamps and subtitle formats still use the deprecated models for existing integrations until shutdown.

Migration checklist

  1. Search for every model name above in code, config and no-code automations.
  2. Do not move from one deprecated model to another. Moving from whisper-1 to gpt-4o-transcribe buys you nothing if both stop on the same day.
  3. Check features you rely on. Word timestamps, speaker labels and subtitle output (SRT/VTT) are exactly the examples the guide ties to the deprecated models. Test them on the replacement or plan another route.
  4. Translation: if you use /v1/audio/translations, it is on the list too. Plan a separate path.
  5. Test on your own audio. Accents, noise and domain vocabulary matter more than any benchmark.
  6. Re-estimate cost. Pricing units differ between models, so the same hour of audio can cost a different amount. Use OpenAI's pricing page: https://developers.openai.com/api/docs/pricing
  7. Set an internal deadline well before 26 February 2027.

Primary sources: the OpenAI speech-to-text guide (https://developers.openai.com/api/docs/guides/speech-to-text) and the OpenAI pricing page above.

For comparing transcription prices across providers on one scale, I keep a Whisper and speech-to-text pricing page with sources and a verification date. Prices change, so confirm on the provider's page.

πŸ“° Read the original article on Dev.to AI

Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes β€” full credit and traffic to the original publisher.