# audio-vad-process

`station__audio__audio-vad-process` · external (needs EXECUTION_BACKEND_URL configured) · domain `audio` · pv-relevance `non-pv`

Process an audio frame through Voice Activity Detection. Returns speech/silence classification with energy, zero-crossing rate, and adaptive threshold.

> **Note:** This tool routes through an external execution backend. If `EXECUTION_BACKEND_URL` is unset on the server, calls return JSON-RPC error `-32603 "Tool execution backend not configured"`. Tools with `backend: native` execute in-process and are always callable.

## Agent metadata

- `idempotent`: unknown
- `read_only`: unknown
- `expected_latency_ms`: unknown (not yet contract-tested)
- `cost_tokens_estimate`: unknown

## Input schema

- `energy_threshold` *number* — Energy threshold for speech detection. Default: 0.02.
- `samples` *array* (required) — F32 audio samples (mono, [-1.0, 1.0]).
- `zcr_ceiling` *number* — Zero-crossing rate ceiling. Frames above this are noise. Default: 0.4.

## Example call

```json
POST /api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "station__audio__audio-vad-process",
    "arguments": {
      "samples": []
    }
  }
}
```

## Related

- [/tools](/tools) — all 3062 tools
- [/tools/audio__audio-vad-process](/tools/audio__audio-vad-process) — HTML page
- [/tools/audio__audio-vad-process/json](/tools/audio__audio-vad-process/json) — JSON form (agent-friendly)
- [/api/mcp](/api/mcp) — endpoint
- [/AGENTS.md](/AGENTS.md) — agent guide