Audio Transcription

Reports audio-transcription decoding capabilities.#

Returns whether ffmpeg is available on the server (so it can decode any format) and the list of file extensions it can currently decode. With ffmpeg the set includes video containers (mp4, mov, mkv, webm, avi, ...): transcription extracts and transcribes their audio track. Clients use this to decide whether to upload a recording as-is or transcode it locally first.

Responses

curl -X GET "$LMKIT_ONE_URL/lmkit/v1/audio-transcription/capabilities" \
  -H "Authorization: Bearer $LMKIT_API_KEY"

Transcribes spoken content from an audio or video file (JSON: base64 or file identifier).#

Accepts an audio or video file as base64 (input_format 'Base64EncodedFile') or as the identifier of a previously uploaded file (input_format 'FileIdentifier'), plus an optional model identifier. 'language' pins the spoken language (ISO-639-1) instead of detecting it; 'enable_vad' controls the voice-activity pre-segmentation that skips silence (on by default), and 'vad' tunes it. Runs speech-to-text on the specified or default model and returns the full transcript and the average confidence score. If processing exceeds the configured timeout, returns 202 Accepted with a job_id. Poll GET /lmkit/v1/jobs/ for status and results.

Video containers (mp4, mov, mkv, webm, avi, ... - see the capabilities route) are accepted when ffmpeg is available on the server: the audio track is extracted and transcribed. A video that declares no audio track answers 422 Unprocessable Content.

Large files: the base64 body is bounded by the configured "Max upload size" (default 100 MB), but base64 inflates a file by ~33%, is buffered whole in memory, and is the least efficient path. For large or long recordings prefer either (a) POST the raw file to /lmkit/v1/audio-transcription/file (multipart, streamed to disk), or (b) POST it once to /lmkit/v1/files/upload and reference the returned fileId here with input_format 'FileIdentifier'. An oversized body is rejected with 413 Payload Too Large.

Request body

application/json · AudioTranscriptionRequest

Responses

curl -X POST "$LMKIT_ONE_URL/lmkit/v1/audio-transcription" \
  -H "Authorization: Bearer $LMKIT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "",
  "input": "",
  "input_format": "Base64EncodedFile",
  "language": "",
  "enable_vad": true
}'

Transcribes spoken content from an uploaded audio or video file (multipart/form-data).#

Accepts an audio or video file as a raw multipart/form-data upload (form field 'file'), plus an optional 'model' form field. This is the recommended path for large or long recordings: the file is streamed to disk rather than buffered whole in memory, and no base64 inflation applies. The upload is bounded by the configured "Max upload size" (default 100 MB); an oversized file is rejected with 413 Payload Too Large. Video containers are accepted when ffmpeg is available on the server: the audio track is extracted and transcribed; a video that declares no audio track answers 422 Unprocessable Content. Behaves identically to the JSON endpoint otherwise: returns the transcript and average confidence, or 202 Accepted with a job_id when processing exceeds the configured timeout (poll GET /lmkit/v1/jobs/).

Request body

application/x-www-form-urlencoded · object

Responses

curl -X POST "$LMKIT_ONE_URL/lmkit/v1/audio-transcription/file" \
  -H "Authorization: Bearer $LMKIT_API_KEY" \
  -H "Content-Type: application/x-www-form-urlencoded"