Enterprise AI Resource Hub - 302.AI | Pay-as-you-go, Comprehensive AI model API Access, Instant Online App Usage

gpt-4o-transcribe

A speech-to-text model from OpenAI

2025-10-31

Audio-Video Processing

Input:

$6/1M tokens

Output:

$10/1M tokens

Bulk order? Contact your manager for exclusive deals

API Overview

GPT-4o-Transcribe is a high-precision speech-to-text model launched by OpenAI, built on the GPT-4o architecture and specifically optimized for speech recognition tasks. The model excels in multilingual, multi-accent, and noisy environments, significantly reducing the word error rate (WER), particularly in English and other major languages. It is ideal for applications such as meeting transcripts, customer service, and media captioning.

API Console

API Analytics

API Reference (1)

API Description	API Endpoint	Request Method	Stability	Parameter Description
Transcriptions（Speech to Text）	https://api.302.ai/v1/audio/transcriptions	POST	Stable	View Details
Document Details OpenAI Guide OpenAI API Reference Transcribe audio into the input language. The Transcription API accepts the audio file you want to transcribe as input, along with your desired output file format for the transcription. We currently support multiple input and output file formats. Price: 0.006 PTC / min Request Parameters Header Parameters Acceptstring Required Example Value:`application/json` Authorizationstring Optional Example Value:`Bearer {{YOUR_API_KEY}}` Request Body multipart/form-data filestring Required The audio file to be transcribed modelenum<string> Required The ID of the model to use. Enum Value: `whisper-1gpt-4o-transcribegpt-4o-mini-transcribegpt-4o-transcribe-diarize` Example Value:`whisper-1` promptstring Optional Optional text to guide the model’s style or continue from a previous audio segment. Prompting guide should match the language of the audio. response_formatenum<string> Optional The format of the transcript output. Options: json. Enum Value: `jsontextsrtverbose_jsonvttdiarized_json` Example Value:`json` temperaturenumber Optional Sampling temperature between 0 and 1. Higher values (e.g., 0.8) make the output more random, while lower values (e.g., 0.2) make it more focused and deterministic. If set to 0, the model automatically increases temperature using log probability until a certain threshold is reached. languagestring Optional The language of the input audio. Providing the input language in ISO-639-1 format will improve accuracy and latency.

API Pricing

$￥円 ₽

Model	Description	Official Price	302.AI Price