MiRA (Music Replication Assessment) tool is a model-independent open evaluation method based on four diverse audio music similarity metrics to assess exact data replication of the training set.

Provide Gradio custom components to make the diarization-based audio labeling process easier and faster.

This repo summarizes the tutorials, datasets, papers, codes and tools for speech separation and speaker extraction task. You are kindly invited to pull requests.

A Real-Recorded and Annotated Microphone Array Dataset for Dynamic Speech Enhancement and Localization

Problems from https://www.deep-ml.com/

Inference code for Audiodec-Valle-Wenetspeech4TTS

Scalable toolkit for data curation

Landing Page for Divide and Remaster v3

PPG-Based Voice Conversion

This repository contains the SpeechBrain Benchmarks

🐟 eza aliases for fish

🌊 The ultimate Fish prompt.

Open implementation of UNIVERSE and UNIVERSE++ diffusion-based speech enhancement models.

Compressed using encodec librilight datasets

target speaker extraction and verification for multi-talker speech

MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra

This is the official implementation of the SEMamba paper.

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

Open-Unmix - Music Source Separation for PyTorch

Spot the conversation: speaker diarisation in the wild

Official repository of the xLSTM.

Codebase for benchmarking several open-sourced SpeechLLM models

Audio time stretch and pitch shift library

Codebase for the paper 'EncodecMAE: Leveraging neural codecs for universal audio representation learning'

Hybrid Demucs model for drum separation

Fetch an academic paper or web article and send it to the reMarkable tablet with a single command