Transcribe audio
carpe_transcribeTranscribes speech from audio URLs, data URIs, or local files (mp3, wav, m4a, flac, ogg, webm) into text using Whisper-compatible ASR with language and vocabulary hints.
Instructions
Speech to text (Whisper-compatible). audio: https:// URL, data: URI, or local path (stdio). Formats such as mp3, wav, m4a, flac, ogg, webm.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"audio": "<audio>"
},
"name": "carpe_transcribe"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| audio | Yes | ||
| model | No | ASR model (default openai/whisper-large-v3). See carpe_list_models type "asr". | |
| route | No | "carpe" → /v1/audio/* (Carpe TEE network). "router" → /router/audio/* (cheapest market). Default from CARPE_DEFAULT_ROUTE. | |
| prompt | No | Context / vocabulary hint. | |
| language | No | ISO-639-1 hint, e.g. "fr". | |
| response_format | No |