← Models

GPT-4o Transcribe

OpenAI · GPT-4o Audio

OpenAI speech-to-text model tier for production transcription and voice pipeline workflows.

Part of GPT family · Other versions: GPT-5.5 Pro , GPT-5.5 , GPT-Rosalind , GPT-5.4 mini , GPT-5.4 nano , GPT-5.4-Pro , GPT-5.4 , GPT-5.3 , GPT-5.3-Codex , GPT-5.2-Codex , GPT-5.2-Pro , GPT-5.2 , GPT-5-Codex , GPT-5 mini , GPT-5 , GPT-5 nano , GPT-4.1 , GPT-4o mini Transcribe , GPT-4o mini TTS , GPT-4o mini , GPT-4o
Type
audio
Context
16K tokens
Max Output
2K tokens
Status
current
API Access
Yes
License
proprietary
speech-to-text transcription audio realtime api
Released March 2025 · Updated May 16, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on May 16, 2026.

GPT-4o Transcribe is OpenAI’s higher-quality speech-to-text model tier for converting spoken audio into text in product and operations workflows. OpenAI’s current model card still presents it as the quality-first transcription route above the mini tier, with better language recognition and lower word error than the original Whisper line.

Capabilities

The model supports high-quality transcription for meeting capture, support workflows, media indexing, and voice-enabled product features. It fits pipelines that need reliable text output from varied audio inputs, especially where accents, noisier clips, or harder audio conditions matter.

Technical Details

OpenAI’s current model docs list a 16K context window and 2K max output tokens for GPT-4o Transcribe. In practice, those numbers matter less than endpoint behavior, file limits, audio quality, and rate limits, but they are useful if you are building around tokenized audio and transcript responses inside the Responses API.

Pricing & Access

OpenAI’s current model docs list GPT-4o Transcribe audio-token pricing at 2.50per1Minputtokensand2.50 per 1M input tokens and 10.00 per 1M output tokens. It is available through OpenAI’s transcription endpoints and broader Responses-style surfaces.

Best Use Cases

Best for transcription services, searchable meeting notes, support call indexing, and ingestion pipelines feeding downstream summarization, QA, or agent workflows.

Comparisons

Compared with GPT-4o mini Transcribe, this tier is positioned for higher quality at roughly double the minute cost. Compared with Whisper, it is the more modern OpenAI route. Internal audio-set testing remains essential because vendor benchmarks rarely reflect your real noise and speaker mix.