OpenAI: gpt-4o-transcribe API
A next-generation speech-to-text model powered by the GPT-4o architecture. It offers significant improvements in Word Error Rate (WER) and multilingual recognition accuracy over Whisper models, excelling in handling thick accents, background noise, and technical jargon.
- Context window: 16,000 tokens
- Max output: 2,000 tokens
- Input: audio, text
- Output: text
- Tool calling: Supported
- File input: Supported
- Released: 2025-03-19
- Knowledge cutoff: 2024-05-31
Frequently Asked Questions
What is the context window of gpt-4o-transcribe?
gpt-4o-transcribe supports a context window of up to 16,000 tokens.
Does gpt-4o-transcribe support function calling?
Yes. gpt-4o-transcribe supports tool / function calling.
What is the knowledge cutoff of gpt-4o-transcribe?
The knowledge cutoff of gpt-4o-transcribe is 2024-05-31.