This repository contains the official code release for Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion.

• Requirements
• Usage Example
  • Text-Based Editing
  • Unsupervised Editing
  • SDEdit
• Citation

python -m pip install -r requirements.txt

Supported models are AudioLDM, TANGO, and AudioLDM2. For unsupervised editing, Stable Diffusion is also supported.

CUDA_VISIBLE_DEVICES=<gpu_num> python main_run.py --cfg_tar <target_cfg_strength> --cfg_src <source_cfg_strength> --init_aud <input_audio_path> --target_prompt <description of the wanted edited signal> --tstart <edit from timestep> --model_id <model_name> --results_path <path to dump results>

You can supply a source prompt that describes the original audio by using --source_prompt.
Use python main_run.py --help for all options.

use --mode ddim to run DDIM inversion and editing (note that --tstart must be equal to num_diffusion_steps (by default set to 200)).

First extract the PCs for your wanted timesteps:

CUDA_VISIBLE_DEVICES=<gpu_num> python main_pc_extract_inv.py  --init_aud <input_audio_path> --model_id <model_name> --results_path <path to dump results> --drift_start <start extraction timestep> --drift_end  <end extraction timestep> --n_evs <amount of evs to extract>

You can supply a source prompt that describes the original audio by using --source_prompt.

Then apply the PCs:

CUDA_VISIBLE_DEVICES=<gpu_num> python main_pc_apply_drift.py --extraction_path <path to extracted .pt file> --drift_start <timestep to start apply> --drift_end <timestep to end apply> --amount <edit strength> --evs <ev nums to apply> 

By using --use_specific_ts_pc <timestep num> you choose a different $t$ from $t'$.
Add --combine_evs to apply all the given PCs together.
Changing --evals_pt to empty will try to get the eigenvalues from the extracted path, and will not work unless the applied timesteps were run in extraction.

Use python main_pc_extract_inv.py --help and python main_pc_apply_drift.py --help for all options.

To recreate the random vectors baseline, use --rand_v. Image samples can be recreated using images_pc_extract_inv.py and images_pc_apply_drift.py.

SDEdit can be run similarly with:

CUDA_VISIBLE_DEVICES=<gpu_num> python main_run_sdedit.py --cfg_tar <target_cfg_strength> --init_aud <input_audio_path> --target_prompt <description of the wanted edited signal> --tstart <edit from timestep> --model_id <model_name> --results_path <path to dump results>

Use python main_run_sdedit.py --help for all options.

Image samples can be recreated using images_run_sdedit.py.

If you use this code for your research, please cite our paper:

@article{manor2024zeroshot,
    title={Zero-Shot Unsupervised and Text-Based Audio Editing Using {DDPM} Inversion}, 
    author={Manor, Hila and Michaeli, Tomer},
    journal={arXiv preprint arXiv:2402.10009},
    year={2024},
}

Parts of this code are heavily based on DDPM Inversion and on Gaussian Denoising Posterior.

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