GPT-4o Transcribe Diarize , a transcription model that identifies who’s speaking when, enables transcripts that clearly associate audio segments with individual speakers. This feature produces the new diarized_json response format, providing you with precise speaker labels along with segment start an…

Saw the model earlier in code pushed yesterday - it’s not been put on the models endpoint yet. [Untitled]

Here is the text from the official announcement email: [ccbfb821549dd6181ef70fbcd475d05ecaa1dcbe_2_1060x1000] Thanks to @multitechvisions for sharing!

Awesome… We’ve been waiting for diarization from OpenAI. But one main missing feature is preventing us from adopting this model. And that’s being able to define the maximum/minimum number of users. We love the known speakers capability, but limiting the users is a crucial feature for us as diarizat…

I did some testing. But the diarization model seems way worse than the old 4o transcribe model. The new model skips over words that the old model didn’t. Anyone else see this?

I’ve had the same experience: the new model completely misses some sentences, which significantly impacts production. The quantitative evaluation results are also worse. Please consider restoring the old model on the gpt-4o-transcribe endpoint.

I tried it multiple times, and it randomly mixes languages…. I’m from Denmark and we transcribe it with danish. ‘language’ is set to ‘da’ as documentation. Anyone else had this issue? One example: We are til gengæld uh mobile when they… The actual danish audio says: Vi er til gengæld øh mobile…

Introducing GPT-4o Transcribe Diarize: Now Available in the Audio API

Community

Andreea_Epure November 4, 2025, 8:58am 13

Yes, it does hallucinate very badly.

Topic		Replies	Views
GPT-4o-transcribe and audio model ready to use via API? API transcribe	10	4227	March 17, 2026
Best solution for Whisper diarization/speaker labeling? API whisper	23	46117	April 9, 2026
How to enable gpt-4o-transcribe-diarize for realtime transcription? API	4	338	February 7, 2026
Audio Models in the API - live stream at 10 AM PT API	15	1269	March 29, 2025
Perhaps realtime API diarization is coming? Documentation api-realtime	3	428	October 21, 2025

Introducing GPT-4o Transcribe Diarize: Now Available in the Audio API

Related topics