# GPT-Realtime-Whisper GPT-Realtime-Whisper is [[OpenAI]]'s streaming [[Speech-to-Text (STT)|speech-to-text]] model, released on 7 May 2026 with [[GPT-Realtime-2]] and [[GPT-Realtime-Translate]]. It transcribes audio as people speak and sends transcript deltas back while the audio is still arriving. Think live captions, meeting notes that keep up with the meeting, or a voice agent that needs to follow the user continuously. Don't confuse it with the original [[Whisper]]. That one is open source (MIT) and runs locally. GPT-Realtime-Whisper is closed and only available through the API, which a [[Hacker News]] commenter called out on launch day. ## Specifications and pricing | | GPT-Realtime-Whisper | |---|---| | Model ID | `gpt-realtime-whisper` | | Endpoint | `v1/realtime/transcription_sessions` | | Input / output | Audio and text in; text out | | Context window | 16,000 tokens | | Max output | 2,000 tokens | | Price | $0.017 per minute of audio | | Rate limits | 100 (tier 1) to 1,300 (tier 5) minutes of audio per minute | The docs describe it as designed for use cases "where developers need to tune latency and accuracy". OpenAI didn't publish word error rates for it in the launch post. ## Superseded as the default On 28 July 2026, OpenAI released `gpt-live-transcribe` (streaming, same $0.017 per minute) and `gpt-transcribe` (files, $0.0045 per minute). The live transcription guide now starts with `gpt-live-transcribe`. GPT-Realtime-Whisper still appears on the pricing page, but it's no longer the first recommendation. On 26 August 2026, OpenAI deprecated `whisper-1`, `gpt-4o-transcribe`, `gpt-4o-mini-transcribe` and `gpt-4o-transcribe-diarize`, with shutdown on 26 February 2027. If you want transcription without sending audio to OpenAI, local models like [[Whisper]], [[Parakeet V3]] or [[Voxtral Transcribe 2]] are the alternatives. ## References - OpenAI, "Advancing voice intelligence with new models in the API": https://openai.com/index/advancing-voice-intelligence-with-new-models-in-the-api/ - Model docs: https://developers.openai.com/api/docs/models/gpt-realtime-whisper - Realtime transcription guide: https://developers.openai.com/api/docs/guides/realtime-transcription - Pricing: https://developers.openai.com/api/docs/pricing - Changelog (gpt-live-transcribe, deprecations): https://developers.openai.com/api/docs/changelog - Hacker News discussion: https://news.ycombinator.com/item?id=48051991 ## Related - [[Whisper]] - [[Speech-to-Text (STT)]] - [[GPT-Realtime-2]] - [[GPT-Realtime-Translate]] - [[OpenAI Realtime API]] - [[2026-05-07 GPT-Realtime-2 - OpenAI's voice model gets GPT-5-class reasoning]]