Introduction
The Speech-to-Text (STT) API based on OpenAI’s Whisper models can convert audio files to text. It supports various use cases:- Transcribing audio files to text
- Translating multilingual audio to English
- Supporting multiple audio format inputs
- Providing multiple output format options
- whisper-large-v3 —— Latest large Whisper model, supports multiple languages. For Chinese recognition, use with appropriate prompts and low temperature values
- whisper-1 —— Original Whisper model, stable and reliable, supports multiple languages
- distil-whisper-large-v3-en —— Distilled model, faster processing speed but slightly lower accuracy, recommended with low temperature values
Model Usage
Speech Transcription
Use/v1/audio/transcriptions endpoint via client.audio.transcriptions.create() method to transcribe audio to text in the original language.
Speech Translation
Use/v1/audio/translations endpoint via client.audio.translations.create() method to translate audio to English text.
Request Parameters
Transcription Parameters
Audio file object to transcribe. Supported formats: mp3, mp4, mpeg, mpga, m4a, wav, webm, maximum 25MB
Language of the input audio in ISO-639-1 format (e.g., ‘en’, ‘zh’). Specifying the language can improve accuracy and latency
Optional text prompt to guide the model’s style or continue a previous audio segment. The prompt should match the audio language
Sampling temperature between 0 and 1. Higher values make output more random, lower values make it more focused and deterministic. Default is 0
Timestamp granularities. Options:
word, segment. Only available when response_format is verbose_jsonTranslation Parameters
Usage Examples
Response Formats
JSON Format (Default)
Verbose JSON Format (verbose_json)
Text Format
SRT Format
VTT Format
Best Practices
- Chinese Audio Processing: Use
whisper-large-v3model, setlanguage="zh",temperature=0.2, and provide appropriate Chinese prompts - English Audio Processing: Use
distil-whisper-large-v3-enfor faster processing speed - Noise Handling: Use prompts to instruct the model to ignore background noise or clean up stammering issues
- Long Audio Processing: API automatically segments long audio; recommend preprocessing audio quality for best results
- Timestamp Requirements: Use
verbose_jsonformat andtimestamp_granularitieswhen precise timestamps are needed - Subtitle Creation: Use
srtorvttformat output directly without additional processing
Last updated: 2026-06-01