amphion/hifigan_speech_bigdata

Amphion Vocoder Pretrained Models

We provide a HiFi-GAN pretrained checkpoint for speech, which is trained on 685 hours of speech data.

Quick Start

To utilize these pretrained vocoders, just run the following commands:

Step1: Download the checkpoint

git lfs install
git clone https://huggingface.co/amphion/hifigan_speech_bigdata

Step2: Clone the Amphion's Source Code of GitHub

git clone https://github.com/open-mmlab/Amphion.git

Step3: Specify the checkpoint's path

Use the soft link to specify the downloaded checkpoint in the first step:

cd Amphion
mkdir -p ckpts/vocoder
ln -s "$(realpath ../hifigan_speech_bigdata/hifigan_speech)" pretrained/hifigan_speech

Step4: Inference

For analysis synthesis on the processed dataset, raw waveform, or predicted mel spectrograms, you can follow the inference part of this recipe.

sh egs/vocoder/gan/tfr_enhanced_hifigan/run.sh --stage 3 \
    --infer_mode [Your chosen inference mode] \
    --infer_datasets [Datasets you want to inference, needed when infer_from_dataset] \
    --infer_feature_dir [Your path to your predicted acoustic features, needed when infer_from_feature] \
    --infer_audio_dir [Your path to your audio files, needed when infer_form_audio] \
    --infer_expt_dir Amphion/ckpts/vocoder/[YourExptName] \
    --infer_output_dir Amphion/ckpts/vocoder/[YourExptName]/result \

Citaions

@misc{gu2023cqt,
      title={Multi-Scale Sub-Band Constant-Q Transform Discriminator for High-Fidelity Vocoder}, 
      author={Yicheng Gu and Xueyao Zhang and Liumeng Xue and Zhizheng Wu},
      year={2023},
      eprint={2311.14957},
      archivePrefix={arXiv},
      primaryClass={cs.SD}
}