Audio
Whisper Speech to Text
Upload speech and download a transcript. Free speech to text on devices you connect.
Whisper by OpenAI; ONNX conversion by ONNX Community
Model licenses
whisper-large-v3-turbo pinned ONNX checkpoint and associated assets · MIT
Retain licence, copyright and publisher attribution notices; no trademark endorsement. Publisher model-card cautions are preserved in the distributed notice.
API reference
Submit via REST
Use your personal API key (msk_…) to create jobs for this workload from your own tools. Jobs started this way run on the same devices you connected on Your Devices — at least one has to be online.
https://cp.meshlabs.cloud/v1/organizations/jobsAuthenticate with X-Organization-Api-Key: msk_your_personal_api_key. The header name matches org keys; personal keys use the same endpoint.
POST https://cp.meshlabs.cloud/v1/organizations/job-inputs/presign with the same API key, upload your file to the returned upload_url, then pass download_url in job_spec (for example as input_image_url).Request body
{
"workload_type": "whisper-large-v3-turbo",
"job_spec": {
"input_audio_url": "https://…/your-input.png",
"language": "en",
"task": "transcribe",
"chunk_seconds": 30
}
}job_spec fields
| Field | Required | Default | Description |
|---|---|---|---|
input_audio_url | Yes | — | Audio to transcribe. Long clips split into 30 s windows. |
language | No | en | Whisper language code. Default English. |
task | No | transcribe | Transcribe in the source language, or translate into English. |
duration_seconds | No | — | Filled from the file probe so the server can fan out 30 s windows. |
chunk_seconds | No | 30 | Workload-constant window. Capped at Whisper's 30 s encoder. |
cURL
curl -X POST https://cp.meshlabs.cloud/v1/organizations/jobs \
-H "X-Organization-Api-Key: msk_your_personal_api_key" \
-H "Content-Type: application/json" \
-d '{
"workload_type": "whisper-large-v3-turbo",
"job_spec": {
"input_audio_url": "https://…/your-input.png",
"language": "en",
"task": "transcribe",
"chunk_seconds": 30
}
}'https://cp.meshlabs.cloud/v1/organizations/jobs/{job_id}Poll with the same API key until status is terminal (COMPLETED, FAILED, CANCELED, or MODERATION_BLOCKED). When content moderation is enabled, jobs may start in MODERATION_PENDING before entering the compute queue. Completed jobs include presigned download_url values on each output.
Create response
{
"id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"workload_type": "whisper-large-v3-turbo",
"status": "PENDING",
"job_spec_json": {
"input_audio_url": "https://…/your-input.png",
"language": "en",
"task": "transcribe",
"chunk_seconds": 30
},
"work_units": [],
"outputs": [],
"created_at": "2026-06-08T12:00:00Z",
"updated_at": "2026-06-08T12:00:00Z"
}Completed response
{
"id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"workload_type": "whisper-large-v3-turbo",
"status": "COMPLETED",
"job_spec_json": {
"input_audio_url": "https://…/your-input.png",
"language": "en",
"task": "transcribe",
"chunk_seconds": 30
},
"work_units": [
{
"id": "…",
"index": 0,
"status": "COMPLETED"
}
],
"outputs": [
{
"id": "…",
"output_type": "<output type>",
"filename": "<output filename>",
"size_bytes": 524288,
"download_url": "<presigned output download URL>"
}
],
"created_at": "2026-06-08T12:00:00Z",
"updated_at": "<completion timestamp>"
}Upload audio and get a transcript from Whisper. Long recordings are split into about 30-second pieces and joined into one transcript. Free speech to text on devices you connect. The model is Apache-2.0. The first run downloads it; later files start faster.