Create transcription
Transcribes audio into text. Accepts base64-encoded audio input as JSON or an OpenAI-style multipart/form-data file upload, and returns the transcribed text.
Authorizations
API key as bearer token in Authorization header
Body
Speech-to-text request input. Accepts a JSON body with input_audio containing base64-encoded audio.
Base64-encoded audio to transcribe
{ "data": "UklGRiQA...", "format": "wav" }
STT model identifier
"openai/whisper-large-v3"
ISO-639-1 language code (e.g., "en", "ja"). Auto-detected if omitted.
"en"
Provider-specific passthrough configuration
Output format. "json" (default) returns { text, usage }. "verbose_json" additionally returns task, language, duration, and segment-level timestamps; only supported by OpenAI-compatible providers.
json, verbose_json "json"
Sampling temperature for transcription
0
Timestamp detail levels to include when response_format is "verbose_json". "segment" returns segment-level timestamps; "word" additionally returns word-level timestamps in the words array. Ignored unless response_format is "verbose_json".
A timestamp detail level for verbose_json transcription responses.
word, segment ["segment"]
Response
Transcription result
STT response containing transcribed text and optional usage statistics
The transcribed text
"Hello, this is a test of OpenAI speech-to-text transcription. The weather is sunny today and the temperature is around 72 degrees."
Duration of the input audio in seconds, present when response_format is verbose_json
9.2
Detected or forced language, present when response_format is verbose_json
"english"
Timestamped transcript segments, present when response_format is verbose_json
The task performed, present when response_format is verbose_json
"transcribe"
Aggregated usage statistics for the request
{
"cost": 0.000508,
"input_tokens": 83,
"output_tokens": 30,
"seconds": 9.2,
"total_tokens": 113
}
Timestamped words, present when the provider returns word-level timestamps