rnnoise_filter

RNNoise noise suppression audio filter for Pipecat.

This module provides an audio filter implementation using RNNoise, a recurrent neural network for audio noise reduction, via the pyrnnoise library.

class pipecat.audio.filters.rnnoise_filter.RNNoiseFilter(resampler_quality: str = 'QQ')

Bases: BaseAudioFilter

Audio filter using RNNoise for noise suppression.

Provides real-time noise suppression for audio streams using RNNoise, a recurrent neural network for audio noise reduction. The filter buffers audio data to match RNNoise’s required frame length (480 samples at 48kHz) and processes it in chunks.

__init__(resampler_quality: str = 'QQ') → None

Initialize the RNNoise noise suppression filter.

Parameters:

resampler_quality – Quality of the resampler if resampling is needed. One of “VHQ”, “HQ”, “MQ”, “LQ”, “QQ”. Defaults to “QQ” (Quick) for lowest latency.

async start(sample_rate: int)

Initialize the filter with the transport’s sample rate.

Parameters:

sample_rate – The sample rate of the input transport in Hz.

async stop()

Clean up the RNNoise engine when stopping.

async process_frame(frame: FilterControlFrame)

Process control frames to enable/disable filtering.

Parameters:

frame – The control frame containing filter commands.

async filter(audio: bytes) → bytes

Apply RNNoise noise suppression to audio data.

Buffers incoming audio and processes it in chunks that match RNNoise’s required frame length (480 samples at 48kHz). Returns filtered audio data.

Parameters:

audio – Raw audio data as bytes to be filtered.

Returns:

Noise-suppressed audio data as bytes.