From ab72ab9577fd9128a6372e7a967baff781ad05e1 Mon Sep 17 00:00:00 2001 From: Yuan_Tuo Date: Tue, 9 Jul 2024 14:34:15 +0800 Subject: [PATCH 1/3] Fix Emilia README.md typo --- preprocessors/Emilia/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/preprocessors/Emilia/README.md b/preprocessors/Emilia/README.md index 63b7975d..12bef64d 100644 --- a/preprocessors/Emilia/README.md +++ b/preprocessors/Emilia/README.md @@ -18,7 +18,7 @@ This is the official repository 👑 for the **Emilia** dataset and the source c Detailed description for the dataset could be found in our paper. -🛠️ **Emilia-Pipe** is the first open-source preprocessing pipeline designed to transform raw, in-the-wild speech data into high-quality training data with annotations for speech generation. This pipeline can process one hour of raw audio into model-ready data in just a few minutes, requiring only the URLs of the audio or video sources. +🛠️ **Emilia-Pipe** is the first open-source preprocessing pipeline designed to transform raw, in-the-wild speech data into high-quality training data with annotations for speech generation. This pipeline can process one hour of raw audio into model-ready data in just a few minutes, requiring only the raw speech data. *To use the Emilia dataset, you can download the raw audio files from the [provided URL list](https://huggingface.co/datasets/amphion/Emilia) and use our open-source [Emilia-Pipe](https://github.com/open-mmlab/Amphion/tree/main/preprocessors/Emilia) preprocessing pipeline to preprocess the raw data and rebuild the dataset. Please note that Emilia doesn't own the copyright of the audios; the copyright remains with the original owners of the video or audio. Additionally, users can easily use Emilia-Pipe to preprocess their own raw speech data for custom needs.* From 81a2abc03fb4d0390409cf4bdd8deaabe2e8457c Mon Sep 17 00:00:00 2001 From: Yuan_Tuo Date: Tue, 9 Jul 2024 14:42:13 +0800 Subject: [PATCH 2/3] Update README.md --- preprocessors/Emilia/README.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/preprocessors/Emilia/README.md b/preprocessors/Emilia/README.md index 12bef64d..e61929b2 100644 --- a/preprocessors/Emilia/README.md +++ b/preprocessors/Emilia/README.md @@ -16,11 +16,13 @@ This is the official repository 👑 for the **Emilia** dataset and the source c - covering six different languages: *English (En), Chinese (Zh), German (De), French (Fr), Japanese (Ja), and Korean (Ko)*; - containing diverse speech data with *various speaking styles*; -Detailed description for the dataset could be found in our paper. +Detailed description for the dataset could be found in our [paper](https://arxiv.org/abs/2407.05361). 🛠️ **Emilia-Pipe** is the first open-source preprocessing pipeline designed to transform raw, in-the-wild speech data into high-quality training data with annotations for speech generation. This pipeline can process one hour of raw audio into model-ready data in just a few minutes, requiring only the raw speech data. -*To use the Emilia dataset, you can download the raw audio files from the [provided URL list](https://huggingface.co/datasets/amphion/Emilia) and use our open-source [Emilia-Pipe](https://github.com/open-mmlab/Amphion/tree/main/preprocessors/Emilia) preprocessing pipeline to preprocess the raw data and rebuild the dataset. Please note that Emilia doesn't own the copyright of the audios; the copyright remains with the original owners of the video or audio. Additionally, users can easily use Emilia-Pipe to preprocess their own raw speech data for custom needs.* +*To use the Emilia dataset, you can download the raw audio files from our provided source URL list on [HuggingFace](https://huggingface.co/datasets/amphion/Emilia) and use our open-source [Emilia-Pipe](.) preprocessing pipeline to preprocess the raw data and rebuild the dataset.* + +*Please note that Emilia doesn't own the copyright of the audios; the copyright remains with the original owners of the video or audio. Additionally, users can easily use Emilia-Pipe to preprocess their own raw speech data for custom needs.* By open-sourcing the Emilia-Pipe code, we aim to enable the speech community to collaborate on large-scale speech generation research. From 2aec4967dfee9b9765e54204ebad20c01945624d Mon Sep 17 00:00:00 2001 From: Yuan_Tuo Date: Tue, 9 Jul 2024 16:38:39 +0800 Subject: [PATCH 3/3] Update README.md --- preprocessors/Emilia/README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/preprocessors/Emilia/README.md b/preprocessors/Emilia/README.md index e61929b2..1b5dd523 100644 --- a/preprocessors/Emilia/README.md +++ b/preprocessors/Emilia/README.md @@ -140,8 +140,8 @@ The processed audio (default 24k sample rate) files will be saved into `input_fo We acknowledge the wonderful work by these excellent developers! - Source Separation: [UVR-MDX-NET-Inst_HQ_3](https://github.com/TRvlvr/model_repo/releases/tag/all_public_uvr_models) - VAD: [snakers4/silero-vad](https://github.com/snakers4/silero-vad) -- Speaker Diarization: [snakers4/silero-vad](https://github.com/snakers4/silero-vad) -- ASR: [m-bain/whisperX](https://github.com/m-bain/whisperX) +- Speaker Diarization: [pyannote/speaker-diarization-3.1](https://huggingface.co/pyannote/speaker-diarization-3.1) +- ASR: [m-bain/whisperX](https://github.com/m-bain/whisperX), using [faster-whisper](https://github.com/guillaumekln/faster-whisper) and [CTranslate2](https://github.com/OpenNMT/CTranslate2) backend. - DNSMOS Prediction: [DNSMOS P.835](https://github.com/microsoft/DNS-Challenge)