arctan-vi 0.7.1

Arctan Voice Isolation for voice agents

Arctan Voice Isolation

Arctan Voice Isolation is a Python SDK for real-time voice isolation in voice agents and audio pipelines.

The SDK supports:

• standard Numpy audio processing
• LiveKit audio input processing
• Pipecat audio input filtering

An Arctan SDK key is required before audio can be processed.

Model and runtime artifacts are downloaded only after authorization and are verified against authorization-provided SHA-256 digests. The encrypted Eigen bundle is decrypted into caller-zeroable process memory and loaded through the runtime's in-memory ABI; SDK 0.7.1 does not materialize plaintext model files, anonymous file descriptors, or named pipes.

Installation

Install the base SDK:

pip install arctan-vi

Install with framework adapters:

pip install "arctan-vi[livekit]"
pip install "arctan-vi[pipecat]"

The base SDK and LiveKit adapter support Python 3.10+. The Pipecat adapter requires Python 3.11+.

Set your SDK key in the runtime environment:

export ARCTAN_SDK_KEY="arc_sk_live_..."

You can also pass license_key="arc_sk_live_..." directly to the processor or adapter constructors. An explicit license_key takes precedence over ARCTAN_SDK_KEY.

Standard Python

For multiple streams, load one shared Model and create one ordered processor per stream. ProcessorAsync runs blocking native inference off the event loop; calls on one processor remain sequential while different processors can run in parallel.

import asyncio
import numpy as np
import arctan

model = arctan.Model()
config = arctan.ProcessorConfig.optimal(model, num_channels=1)
call_a = arctan.ProcessorAsync(model=model, config=config)
call_b = arctan.ProcessorAsync(
    model=model,
    config=arctan.ProcessorConfig.optimal(model),
)

async def process_calls():
    audio = np.zeros((config.num_channels, config.num_frames), dtype=np.float32)
    enhanced_a, enhanced_b = await asyncio.gather(
        call_a.process_async(audio),
        call_b.process_async(audio),
    )
    return enhanced_a, enhanced_b

try:
    enhanced_a, enhanced_b = asyncio.run(process_calls())
finally:
    call_a.close()
    call_b.close()
    model.close()

Create exactly one processor per audio stream and do not submit frames from the same stream out of order. The number of processors is the number of active streams; it is not a universal host-capacity claim.

The existing single-stream convenience remains supported. It creates and owns a private model internally:

processor = arctan.Processor(
    config=arctan.ProcessorConfig(sample_rate=24000),
)

Processor.process() accepts float32 audio shaped as channels x frames. Samples must be finite and in [-1.0, 1.0]. The output is a new float32 array with the same shape.

Mono and stereo input are supported. Stereo input is downmixed for voice isolation and returned as dual-mono stereo. The sample rate must remain stable for the lifetime of a processor.

If num_frames is omitted, initialization fills it with the native frame size for the selected sample rate. If you set num_frames, it must contain a whole number of native processing blocks.

LiveKit

Install the LiveKit extra:

pip install "arctan-vi[livekit]"

Create an Arctan noise-cancellation processor and pass it to LiveKit audio input options:

from arctan.livekit import arctan_enhancer
from livekit.agents import room_io

await session.start(
    agent=Assistant(),
    room=ctx.room,
    room_options=room_io.RoomOptions(
        audio_input=room_io.AudioInputOptions(
            noise_cancellation=arctan_enhancer.noise_canceller(),
        ),
    ),
)

You can pass SDK options directly to Arctan:

noise_cancellation = arctan_enhancer.noise_canceller(
    license_key="arc_sk_live_...",
)

LiveKit mono and stereo frames are supported. Stereo frames are downmixed for voice isolation and returned as dual-mono stereo. The sample rate must remain stable, and each input frame must contain a whole number of native processing blocks.

Pipecat

Install the Pipecat extra:

pip install "arctan-vi[pipecat]"

Create an ArctanAudioFilter and pass it as the input audio filter for your transport:

from arctan.pipecat import arctan_enhancer
from pipecat.transports.base_transport import TransportParams

audio_filter = arctan_enhancer.ArctanAudioFilter()

transport_params = TransportParams(
    audio_in_enabled=True,
    audio_in_filter=audio_filter,
)

ArctanAudioFilter implements Pipecat's BaseAudioFilter. Pipecat calls start(sample_rate), filter(audio), and stop() through the transport lifecycle.

The Pipecat adapter accepts mono PCM16 input. Arbitrary byte chunks are buffered until a complete native processing block is available, so filter() may return b"". FilterEnableFrame can bypass or re-enable processing; toggling clears any incomplete buffered block.