> ## Documentation Index
> Fetch the complete documentation index at: https://doc.rapida.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Voice Activity Detection — Overview

> VAD interface, factory function, providers, and model setup in assistant-api.

Voice Activity Detection (VAD) runs on every audio frame before STT. It determines whether the caller is speaking, emits interruption signals for barge-in, and sends speech activity heartbeats that keep the End of Speech timer from firing during active speech.

All three VAD providers run locally using ONNX or native C libraries. No external API calls, no credentials, no added latency.

***

## VAD Interface

Every provider implements the `Vad` interface:

```go theme={null}
// api/assistant-api/internal/type/vad.go
type Vad interface {
    Name() string
    Process(ctx context.Context, pkt UserAudioPacket) error
    Close() error
}
```

Input audio is always **16 kHz LINEAR16 mono** — the platform's internal format. Resampling from 8 kHz telephony audio happens upstream before VAD sees it.

***

## Factory Function

```go theme={null}
// api/assistant-api/internal/vad/vad.go
func GetVAD(ctx, logger, callback, options) (Vad, error)
```

The factory reads `microphone.vad.provider` from the assistant's audio options and returns the matching implementation. If no provider is set, **Silero VAD** is used as the default.

***

## Provider Identifiers

| Identifier    | Provider    | Model                                            |
| ------------- | ----------- | ------------------------------------------------ |
| `silero_vad`  | Silero VAD  | `silero_vad_20251001.onnx` (\~2 MB)              |
| `ten_vad`     | TEN VAD     | Native C library (no model file)                 |
| `firered_vad` | FireRed VAD | `fireredvad_stream_vad_with_cache.onnx` (\~5 MB) |

***

## Shared Parameters

All three providers use the same configuration keys:

| Option Key                         | Type   | Default      | Description                                                                                         |
| ---------------------------------- | ------ | ------------ | --------------------------------------------------------------------------------------------------- |
| `microphone.vad.provider`          | string | `silero_vad` | Provider selection                                                                                  |
| `microphone.vad.threshold`         | float  | `0.5`        | Speech probability threshold (0.3–1.0). Lower = more sensitive.                                     |
| `microphone.vad.min_silence_frame` | float  | `20`         | Consecutive silence frames before ending a speech segment. Each frame = 10 ms. Default 20 = 200 ms. |
| `microphone.vad.min_speech_frame`  | float  | `8`          | Consecutive speech frames before confirming speech onset. Each frame = 10 ms. Default 8 = 80 ms.    |

***

## Model Files

VAD models are bundled in the Docker image at build time. For source builds, models are resolved from the source tree relative to each provider's Go package directory.

### Docker

Models are copied into the runtime image and referenced via environment variables:

```env theme={null}
SILERO_MODEL_PATH=./models/silero_vad/silero_vad_20251001.onnx
FIRERED_VAD_MODEL_PATH=./models/firered_vad/fireredvad_stream_vad_with_cache.onnx
```

TEN VAD uses a native shared library instead of an ONNX model:

```env theme={null}
LD_LIBRARY_PATH="/opt/ten_vad/lib:..."
```

These are set automatically in the Dockerfile — no manual configuration needed when running via Docker Compose.

### From Source

When running `go run cmd/assistant/assistant.go`, each provider resolves its model path using `runtime.Caller` to find the source directory:

| Provider    | Default path (relative to package dir)                                                             |
| ----------- | -------------------------------------------------------------------------------------------------- |
| Silero VAD  | `api/assistant-api/internal/vad/internal/silero_vad/models/silero_vad_20251001.onnx`               |
| FireRed VAD | `api/assistant-api/internal/vad/internal/firered_vad/models/fireredvad_stream_vad_with_cache.onnx` |
| TEN VAD     | No model file — requires `libten_vad.so` in the library path                                       |

Silero and FireRed models are checked into the repository. TEN VAD requires the shared library to be installed or available in `LD_LIBRARY_PATH` / `DYLIB_LIBRARY_PATH`.

To override model paths, set the environment variables:

```bash theme={null}
export SILERO_MODEL_PATH=/path/to/silero_vad.onnx
export FIRERED_VAD_MODEL_PATH=/path/to/fireredvad.onnx
```

***

## CGO Dependencies

All VAD providers use CGO for inference:

| Provider    | CGO dependency                  | Build flags                                                       |
| ----------- | ------------------------------- | ----------------------------------------------------------------- |
| Silero VAD  | ONNX Runtime (`libonnxruntime`) | `-I/opt/onnxruntime/include -L/opt/onnxruntime/lib -lonnxruntime` |
| FireRed VAD | ONNX Runtime (`libonnxruntime`) | Same as Silero                                                    |
| TEN VAD     | TEN VAD library (`libten_vad`)  | `-I/opt/ten_vad/include -L/opt/ten_vad/lib -lten_vad`             |

The Docker base image (`rapidaai/rapida-golang`) includes all CGO dependencies pre-installed. For local source builds, you need ONNX Runtime and the TEN VAD shared library installed on your system.

***

## Providers

| Provider                                                          | Approach                           | Best for                               |
| ----------------------------------------------------------------- | ---------------------------------- | -------------------------------------- |
| **[Silero VAD](/opensource/services/assistant-api/vad/silero)**   | ONNX model, 100+ languages         | General purpose — the default          |
| **[TEN VAD](/opensource/services/assistant-api/vad/ten)**         | Native C library, 16 ms frame hops | Lowest per-frame latency               |
| **[FireRed VAD](/opensource/services/assistant-api/vad/firered)** | DFSMN model, 4-state postprocessor | Noisy environments, precise boundaries |

See the [VAD concepts guide](/assistants/voice-activity-detection) for detailed parameter tuning guidance and use-case recommendations.