Speech Enhancement Python Packages

deepfilternet

Noise supression using deep filtering

espnet

End-to-End Speech Processing Toolkit

pvkoala

On-device noise suppression powered by deep learning

asteroid

The PyTorch-based audio source separation toolkit for researchers

clearvoice

An AI-Powered Speech Processing Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Enhancement, Separation, and Target Speaker Extraction, etc.

resemble-enhance

AI powered speech denoising and enhancement

torch-pesq

PyTorch implementation of the Perceptual Evaluation of Speech Quality for wideband audio

pyclarity

Clarity Challenge toolkit - software for building Clarity Challenge systems

voicefixer

General Speech Restoration

dpdfnet

Clean up noisy speech in real time with DPDFNet - open-source streaming speech enhancement for research, audio apps, and edge devices. Includes pretrained models, PyTorch code, ONNX/TFLite inference, 8/16/48 kHz support, and live demos.

rapidly

Realtime audio separation running fully on-device across iOS, macOS, Android, Linux and Windows

denoisers

Simple PyTorch Denoisers for Waveform Audio

textual-echo-cancellation

Textual Echo Cancellation

speechdsp

Speech DSP toolkit in pure NumPy/SciPy: STFT, MFCC, VAD, log-MMSE enhancement and imbalance-aware metrics (UAR). No librosa or deep learning required.

mpsenet

Python package of MP-SENet from Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancement.

pvkoalademo

On-device noise suppression powered by deep learning

speech-selib

🎙️ A small speech enhancement library using ONNX models 🚀📦

torchsubband

Pytorch implementation of subband decomposition

sprocket-vc

Voice Conversion Tool Kit

openvox

OpenVox is a complete voice stack speech-to-text and text-to-speech that runs entirely on your own hardware, with no cloud, no API keys, and no per-minute fees. It's built for the places cloud voice services can't go: robots, embedded and edge devices, air-gapped systems, and any product where audio must never leave the machine.

cleanvoice

A Fast Speech Enhancement toolkit using Conv-TasNet

deepfilternet-fix

Noise supression using deep filtering

nectarstt

OpenVox is a complete voice stack speech-to-text and text-to-speech that runs entirely on your own hardware, with no cloud, no API keys, and no per-minute fees. It's built for the places cloud voice services can't go: robots, embedded and edge devices, air-gapped systems, and any product where audio must never leave the machine.

phonepod

Local AI audio restoration. Phone recording to podcast quality. Zero cloud.