LiveKit Transport Provider

Not yet functional

The livekit provider is registered but not implemented: it never dials LiveKit. transport.New("livekit", ...) and livekit.New always return an error wrapping transport.ErrNotImplemented (which also wraps errors.ErrUnsupported), and every method on a livekit.Transport value returns the same error instead of silently discarding audio. The configuration and examples below describe the planned integration. Use transport.NewWebSocketTransport for a working transport today. Track progress in the framework repository.

The LiveKit provider implements bidirectional audio transport through LiveKit rooms. LiveKit is treated as a transport layer — it provides WebRTC connectivity while Beluga handles all STT, LLM, and TTS processing through the frame-based pipeline.

Choose LiveKit when you need a self-hostable, open-source WebRTC infrastructure for production voice deployments. LiveKit provides room management, access token authentication, and scalable media routing. Beluga treats LiveKit as a pure transport layer, handling all voice processing in the frame-based pipeline. For a managed alternative, consider Daily.

Installation

go get github.com/lookatitude/beluga-ai/v2/voice/transport/providers/livekit

Configuration

Field	Required	Default	Description
URL	Yes	—	LiveKit server URL (wss://...)
Token	Yes	—	LiveKit access token
SampleRate	No	16000	Audio sample rate in Hz
Channels	No	1	Audio channels (1=mono, 2=stereo)

Extra configuration:

Key	Type	Description
room	string	LiveKit room name to join

Basic Usage

package main

import (
    "context"
    "fmt"
    "log"

    "github.com/lookatitude/beluga-ai/v2/voice/transport"
    _ "github.com/lookatitude/beluga-ai/v2/voice/transport/providers/livekit"
)

func main() {
    t, err := transport.New("livekit", transport.Config{
        URL:   "wss://myapp.livekit.cloud",
        Token: "your-livekit-token",
        Extra: map[string]any{
            "room": "voice-room",
        },
    })
    if err != nil {
        log.Fatal(err)
    }
    defer t.Close()

    // Receive incoming audio frames
    for frame, err := range t.Recv(context.Background()) {
        if err != nil {
            log.Fatal(err)
        }
        fmt.Printf("Received %d bytes of audio\n", len(frame.Data))
    }
}

Advanced Features

Sending Audio

Write processed audio back to the LiveKit room using voice.NewAudioFrame:

frame := voice.NewAudioFrame(processedPCM, 16000)
err := t.Send(ctx, frame)
if err != nil {
    log.Printf("send failed: %v", err)
}

Room Configuration

Pass the room name through the Extra map:

t, err := transport.New("livekit", transport.Config{
    URL:   "wss://myapp.livekit.cloud",
    Token: token,
    Extra: map[string]any{
        "room": "my-voice-room",
    },
})

Voice Pipeline Integration

Wrap the transport for use with Beluga’s voice pipeline:

import "github.com/lookatitude/beluga-ai/v2/voice/transport"

audioTransport, err := transport.New("livekit", cfg)
if err != nil {
    log.Fatal(err)
}

voiceTransport := &transport.AsVoiceTransport{T: audioTransport}
// Use voiceTransport with the voice pipeline

Direct Construction

import "github.com/lookatitude/beluga-ai/v2/voice/transport/providers/livekit"

t, err := livekit.New(transport.Config{
    URL:        "wss://myapp.livekit.cloud",
    Token:      "your-livekit-token",
    SampleRate: 16000,
    Channels:   1,
    Extra:      map[string]any{"room": "voice-room"},
})

Error Handling

transport.New("livekit", ...) currently always fails with an error wrapping transport.ErrNotImplemented, so check for it and fall back to a working transport:

t, err := transport.New("livekit", cfg)
if errors.Is(err, transport.ErrNotImplemented) {
    // livekit is not implemented yet; use the built-in WebSocket transport
    t, err = transport.NewWebSocketTransport(ctx, wsURL)
}
if err != nil {
    log.Fatal(err)
}
defer t.Close()

Refer to the LiveKit documentation for server setup, room management, and token generation.