百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
A

Apollo

> 编程语言
开源

音乐修复方法可将损失性的 MP3 压缩音乐转换为无损音乐。

396 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

音乐修复方法可将损失性的 MP3 压缩音乐转换为无损音乐。

Kai Li1,2, Yi Luo2

1Tsinghua University, Beijing, China

2Tencent AI Lab, Shenzhen, China

ArXiv | Demo

Apollo: Band-sequence Modeling for High-Quality Audio Restoration

English | Traditional Chinese guide

Abstract

Audio restoration has become increasingly significant in modern society, not only due to the demand for high-quality auditory experiences enabled by advanced playback devices, but also because the growing capabilities of generative audio models necessitate high-fidelity audio. Typically, audio restoration is defined as a task of predicting undistorted audio from damaged input, often trained using a GAN framework to balance perception and distortion. Since audio degradation is primarily concentrated in mid- and high-frequency ranges, especially due to codecs, a key challenge lies in designing a generator capable of preserving low-frequency information while accurately reconstructing high-quality mid- and high-frequency content. Inspired by recent advancements in high-sample-rate music separation, speech enhancement, and audio codec models, we propose Apollo, a generative model designed for high-sample-rate audio restoration. Apollo employs an explicit frequency band split module to model the relationships between different frequency bands, allowing for more coherent and higher-quality restored audio. Evaluated on the MUSDB18-HQ and MoisesDB datasets, Apollo consistently outperforms existing SR-GAN models across various bit rates and music genres, particularly excelling in complex scenarios involving mixtures of multiple instruments and vocals. Apollo significantly improves music restoration quality while maintaining computational efficiency.

News

  • [2025.03.07] We released the training data preprocessing code on Apollo-data-preprocess.
  • [2024.09.10] Apollo is now available on ArXiv and Demo.
  • [2024.09.10] Apollo checkpoints and pre-trained models are available for download.

⚡️ Installation

Clone the repository

bash
git clone https://github.com/JusperLee/Apollo.git && cd Apollo
conda create --name look2hear --file look2hear.yml
conda activate look2hear

️ Usage

️ Datasets

Apollo is trained on the MUSDB18-HQ and MoisesDB datasets. To download the datasets, run the following commands:

bash
wget https://zenodo.org/records/3338373/files/musdb18hq.zip?download=1
wget https://ds-website-downloads.55c2710389d9da776875002a7d018e59.r2.cloudflarestorage.com/moisesdb.zip

During data preprocessing, we drew inspiration from music separation techniques and implemented the following steps:

  1. Source Activity Detection (SAD):
    We used a Source Activity Detector (SAD) to remove silent regions from the audio tracks, retaining only the significant portions for training.

  2. Data Augmentation:
    We performed real-time data augmentation by mixing tracks from different songs. For each mix, we randomly selected between 1 and 8 stems from the 11 available tracks, extracting 3-second clips from each selected stem. These clips were scaled in energy by a random factor within the range of [-10, 10] dB relative to their original levels. The selected clips were then summed together to create simulated mixed music.

  3. Simulating Dynamic Bitrate Compression:
    We simulated various bitrate scenarios by applying MP3 codecs with bitrates of [24000, 32000, 48000, 64000, 96000, 128000].

  4. Rescaling:
    To ensure consistency across all samples, we rescaled both the target and the encoded audio based on their maximum absolute values.

  5. Saving as HDF5:
    After preprocessing, all data (including the source stems, mixed tracks, and compressed audio) was saved in HDF5 format, making it easy to load for training and evaluation purposes.

Training

To train the Apollo model, run the following command:

bash
python train.py --conf_dir=configs/apollo.yml

Evaluation

To evaluate the Apollo model, run the following command:

bash
python inference.py \
  --in_wav=asserts/input_wav.wav \
  --out_wav=output.wav \
  --device=auto

--device=auto prefers CUDA, then Apple Silicon MPS, and finally CPU. The default downloads the official JusperLee/Apollo checkpoint from Hugging Face. For a custom checkpoint, --checkpoint accepts only an existing local file; arbitrary remote repositories are rejected because the upstream checkpoint loader deserializes the file with PyTorch.

Full-file inference remains the default. For long audio, enable bounded-memory inference explicitly:

bash
python inference.py \
  --in_wav=long_input.wav \
  --out_wav=long_output.wav \
  --device=auto \
  --chunk-seconds=6 \
  --overlap-seconds=1 \
  --chunk-pad-seconds=1 \
  --chunk-batch-size=2

--chunk-batch-size groups a fixed number of chunks into each model forward pass. Its default is 1 for minimum accelerator memory; larger values may improve throughput on some devices but use more memory and should be measured on the target backend. Normalized linear crossfades preserve the input frame count, and the overlap must be no more than half the chunk duration. --chunk-pad-seconds infers each chunk with that much surrounding audio per side and discards it from the output; with its default of 0 the listener may hear artifacts at the chunk seams. Chunking can change the result relative to full-file inference because the model sees less context, so important material should still receive a listening check. See Apple Silicon macOS notes for a tested MPS setup and current limitations.

Results

Here, you can include a brief overview of the performance metrics or results that Apollo achieves using different bitrates

Different methods' SDR/SI-SNR/VISQOL scores for various types of music, as well as the number of model parameters and GPU inference time. For the GPU inference time test, a music signal with a sampling rate of 44.1 kHz and a length of 1 second was used.

License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Third Party

Apollo-Colab-Inference

Acknowledgements

Apollo is developed by the Look2Hear at Tsinghua University.

Citation

If you use Apollo in your research or project, please cite the following paper:

bibtex
@inproceedings{li2025apollo,
  title={Apollo: Band-sequence Modeling for High-Quality Music Restoration in Compressed Audio},
  author={Li, Kai and Luo, Yi},
  booktitle={IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2025},
  organization={IEEE}
}

Contact

For any questions or feedback regarding Apollo, feel free to reach out to us via email: [email protected]

Issues· 29 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

Python

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月17日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言