lucataco/zeta-editing – Run with an API on Replicate

lucataco / zeta-editing

Zero-Shot Text-Based Audio Editing Using DDPM Inversion

Cold

Public
1.8K runs
L40S
GitHub
Paper
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

Video Player is loading.

Current Time 00:00:000

Duration 00:00:000

Loaded: 0%

Stream Type LIVE

Remaining Time 00:00:000

audio

*file

Input Audio File

prompt

string

Shift + Return to add a new line

A recording of an arcade game soundtrackA recording of an arcade game soundtrack

Describe your desired edited output

Default: "A recording of an arcade game soundtrack"

t_start

integer

(minimum: 15, maximum: 85)

Lower % returns closer to the original audio, higher returns stronger edit

Default: 45

audio_version

string

Choose the audio version to return

Default: "cvssp/audioldm2-music"

source_prompt

string

Shift + Return to add a new line

Optional: describe the original audio input

Default: ""

steps

integer

Number of diffusion steps, higher values(200) yield high-quality generations

Default: 50

cfg_scale_src

number

Source Guidance Scale

Default: 3

cfg_scale_tar

number

Target Guidance Scale

Default: 12

seed

integer

Random seed

Run this model in Node.js with one line of code:

npx create-replicate --model=lucataco/zeta-editing

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/zeta-editing using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/zeta-editing:ff80c3cca3dc792e2cb119367dc11295b499cda961a2d0b34ecf5109b59c6528",
  {
    input: {
      audio: "https://replicate.delivery/pbxt/KVudHMNiL8E0LcDPJfRr8WzgmY5Mry4d2uhYqMyU7xseJdWr/Beethoven.wav",
      steps: 50,
      prompt: "A recording of an arcade game soundtrack",
      t_start: 45,
      audio_version: "cvssp/audioldm2-music",
      cfg_scale_src: 3,
      cfg_scale_tar: 12,
      source_prompt: ""
    }
  }
);

// To access the file URL:
console.log(output.url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run lucataco/zeta-editing using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/zeta-editing:ff80c3cca3dc792e2cb119367dc11295b499cda961a2d0b34ecf5109b59c6528",
    input={
        "audio": "https://replicate.delivery/pbxt/KVudHMNiL8E0LcDPJfRr8WzgmY5Mry4d2uhYqMyU7xseJdWr/Beethoven.wav",
        "steps": 50,
        "prompt": "A recording of an arcade game soundtrack",
        "t_start": 45,
        "audio_version": "cvssp/audioldm2-music",
        "cfg_scale_src": 3,
        "cfg_scale_tar": 12,
        "source_prompt": ""
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run lucataco/zeta-editing using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lucataco/zeta-editing:ff80c3cca3dc792e2cb119367dc11295b499cda961a2d0b34ecf5109b59c6528",
    "input": {
      "audio": "https://replicate.delivery/pbxt/KVudHMNiL8E0LcDPJfRr8WzgmY5Mry4d2uhYqMyU7xseJdWr/Beethoven.wav",
      "steps": 50,
      "prompt": "A recording of an arcade game soundtrack",
      "t_start": 45,
      "audio_version": "cvssp/audioldm2-music",
      "cfg_scale_src": 3,
      "cfg_scale_tar": 12,
      "source_prompt": ""
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

Video Player is loading.

Current Time 00:00:000

Duration 00:00:000

Loaded: 0%

Stream Type LIVE

Remaining Time 00:00:000

{
  "completed_at": "2024-03-05T00:03:23.912276Z",
  "created_at": "2024-03-05T00:03:08.892754Z",
  "data_removed": false,
  "error": null,
  "id": "e2chgctbequgag37qvizvrvkym",
  "input": {
    "audio": "https://replicate.delivery/pbxt/KVudHMNiL8E0LcDPJfRr8WzgmY5Mry4d2uhYqMyU7xseJdWr/Beethoven.wav",
    "steps": 50,
    "prompt": "A recording of an arcade game soundtrack",
    "t_start": 45,
    "audio_version": "cvssp/audioldm2-music",
    "cfg_scale_src": 3,
    "cfg_scale_tar": 12,
    "source_prompt": ""
  },
  "logs": "Using seed: 1060458061\nUsing model: cvssp/audioldm2-music\n  0%|          | 0/50 [00:00<?, ?it/s]\n  4%|▍         | 2/50 [00:00<00:03, 12.68it/s]\n  8%|▊         | 4/50 [00:00<00:03, 12.36it/s]\n 12%|█▏        | 6/50 [00:00<00:03, 12.40it/s]\n 16%|█▌        | 8/50 [00:00<00:03, 12.40it/s]\n 20%|██        | 10/50 [00:00<00:03, 12.58it/s]\n 24%|██▍       | 12/50 [00:00<00:03, 12.62it/s]\n 28%|██▊       | 14/50 [00:01<00:02, 12.67it/s]\n 32%|███▏      | 16/50 [00:01<00:02, 12.73it/s]\n 36%|███▌      | 18/50 [00:01<00:02, 12.68it/s]\n 40%|████      | 20/50 [00:01<00:02, 12.67it/s]\n 44%|████▍     | 22/50 [00:01<00:02, 12.57it/s]\n 48%|████▊     | 24/50 [00:01<00:02, 12.56it/s]\n 52%|█████▏    | 26/50 [00:02<00:01, 12.64it/s]\n 56%|█████▌    | 28/50 [00:02<00:01, 12.72it/s]\n 60%|██████    | 30/50 [00:02<00:01, 12.68it/s]\n 64%|██████▍   | 32/50 [00:02<00:01, 12.61it/s]\n 68%|██████▊   | 34/50 [00:02<00:01, 12.50it/s]\n 72%|███████▏  | 36/50 [00:02<00:01, 12.56it/s]\n 76%|███████▌  | 38/50 [00:03<00:00, 12.62it/s]\n 80%|████████  | 40/50 [00:03<00:00, 12.62it/s]\n 84%|████████▍ | 42/50 [00:03<00:00, 12.58it/s]\n 88%|████████▊ | 44/50 [00:03<00:00, 12.60it/s]\n 92%|█████████▏| 46/50 [00:03<00:00, 12.62it/s]\n 96%|█████████▌| 48/50 [00:03<00:00, 12.63it/s]\n100%|██████████| 50/50 [00:03<00:00, 12.67it/s]\n100%|██████████| 50/50 [00:03<00:00, 12.61it/s]\n  0%|          | 0/22 [00:00<?, ?it/s]\n  5%|▍         | 1/22 [00:00<00:03,  6.44it/s]\n  9%|▉         | 2/22 [00:00<00:03,  6.39it/s]\n 14%|█▎        | 3/22 [00:00<00:02,  6.39it/s]\n 18%|█▊        | 4/22 [00:00<00:02,  6.44it/s]\n 23%|██▎       | 5/22 [00:00<00:02,  6.44it/s]\n 27%|██▋       | 6/22 [00:00<00:02,  6.47it/s]\n 32%|███▏      | 7/22 [00:01<00:02,  6.46it/s]\n 36%|███▋      | 8/22 [00:01<00:02,  6.47it/s]\n 41%|████      | 9/22 [00:01<00:02,  6.44it/s]\n 45%|████▌     | 10/22 [00:01<00:01,  6.43it/s]\n 50%|█████     | 11/22 [00:01<00:01,  6.40it/s]\n 55%|█████▍    | 12/22 [00:01<00:01,  6.42it/s]\n 59%|█████▉    | 13/22 [00:02<00:01,  6.44it/s]\n 64%|██████▎   | 14/22 [00:02<00:01,  6.46it/s]\n 68%|██████▊   | 15/22 [00:02<00:01,  6.47it/s]\n 73%|███████▎  | 16/22 [00:02<00:00,  6.44it/s]\n 77%|███████▋  | 17/22 [00:02<00:00,  6.43it/s]\n 82%|████████▏ | 18/22 [00:02<00:00,  6.41it/s]\n 86%|████████▋ | 19/22 [00:02<00:00,  6.44it/s]\n 91%|█████████ | 20/22 [00:03<00:00,  6.45it/s]\n 95%|█████████▌| 21/22 [00:03<00:00,  6.44it/s]\n100%|██████████| 22/22 [00:03<00:00,  6.46it/s]\n100%|██████████| 22/22 [00:03<00:00,  6.44it/s]",
  "metrics": {
    "predict_time": 14.98052,
    "total_time": 15.019522
  },
  "output": "https://replicate.delivery/pbxt/lZaeHeSf9zJ7ZILM52Upzy2aZQbKm4xvbzAbtwyux5xKaC6kA/tmpdlk029zc.wav",
  "started_at": "2024-03-05T00:03:08.931756Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/e2chgctbequgag37qvizvrvkym",
    "cancel": "https://api.replicate.com/v1/predictions/e2chgctbequgag37qvizvrvkym/cancel"
  },
  "version": "ff80c3cca3dc792e2cb119367dc11295b499cda961a2d0b34ecf5109b59c6528"
}

Generated in

15.0 seconds

Tweak itReport

Using seed: 1060458061
Using model: cvssp/audioldm2-music
  0%|          | 0/50 [00:00<?, ?it/s]
  4%|▍         | 2/50 [00:00<00:03, 12.68it/s]
  8%|▊         | 4/50 [00:00<00:03, 12.36it/s]
 12%|█▏        | 6/50 [00:00<00:03, 12.40it/s]
 16%|█▌        | 8/50 [00:00<00:03, 12.40it/s]
 20%|██        | 10/50 [00:00<00:03, 12.58it/s]
 24%|██▍       | 12/50 [00:00<00:03, 12.62it/s]
 28%|██▊       | 14/50 [00:01<00:02, 12.67it/s]
 32%|███▏      | 16/50 [00:01<00:02, 12.73it/s]
 36%|███▌      | 18/50 [00:01<00:02, 12.68it/s]
 40%|████      | 20/50 [00:01<00:02, 12.67it/s]
 44%|████▍     | 22/50 [00:01<00:02, 12.57it/s]
 48%|████▊     | 24/50 [00:01<00:02, 12.56it/s]
 52%|█████▏    | 26/50 [00:02<00:01, 12.64it/s]
 56%|█████▌    | 28/50 [00:02<00:01, 12.72it/s]
 60%|██████    | 30/50 [00:02<00:01, 12.68it/s]
 64%|██████▍   | 32/50 [00:02<00:01, 12.61it/s]
 68%|██████▊   | 34/50 [00:02<00:01, 12.50it/s]
 72%|███████▏  | 36/50 [00:02<00:01, 12.56it/s]
 76%|███████▌  | 38/50 [00:03<00:00, 12.62it/s]
 80%|████████  | 40/50 [00:03<00:00, 12.62it/s]
 84%|████████▍ | 42/50 [00:03<00:00, 12.58it/s]
 88%|████████▊ | 44/50 [00:03<00:00, 12.60it/s]
 92%|█████████▏| 46/50 [00:03<00:00, 12.62it/s]
 96%|█████████▌| 48/50 [00:03<00:00, 12.63it/s]
100%|██████████| 50/50 [00:03<00:00, 12.67it/s]
100%|██████████| 50/50 [00:03<00:00, 12.61it/s]
  0%|          | 0/22 [00:00<?, ?it/s]
  5%|▍         | 1/22 [00:00<00:03,  6.44it/s]
  9%|▉         | 2/22 [00:00<00:03,  6.39it/s]
 14%|█▎        | 3/22 [00:00<00:02,  6.39it/s]
 18%|█▊        | 4/22 [00:00<00:02,  6.44it/s]
 23%|██▎       | 5/22 [00:00<00:02,  6.44it/s]
 27%|██▋       | 6/22 [00:00<00:02,  6.47it/s]
 32%|███▏      | 7/22 [00:01<00:02,  6.46it/s]
 36%|███▋      | 8/22 [00:01<00:02,  6.47it/s]
 41%|████      | 9/22 [00:01<00:02,  6.44it/s]
 45%|████▌     | 10/22 [00:01<00:01,  6.43it/s]
 50%|█████     | 11/22 [00:01<00:01,  6.40it/s]
 55%|█████▍    | 12/22 [00:01<00:01,  6.42it/s]
 59%|█████▉    | 13/22 [00:02<00:01,  6.44it/s]
 64%|██████▎   | 14/22 [00:02<00:01,  6.46it/s]
 68%|██████▊   | 15/22 [00:02<00:01,  6.47it/s]
 73%|███████▎  | 16/22 [00:02<00:00,  6.44it/s]
 77%|███████▋  | 17/22 [00:02<00:00,  6.43it/s]
 82%|████████▏ | 18/22 [00:02<00:00,  6.41it/s]
 86%|████████▋ | 19/22 [00:02<00:00,  6.44it/s]
 91%|█████████ | 20/22 [00:03<00:00,  6.45it/s]
 95%|█████████▌| 21/22 [00:03<00:00,  6.44it/s]
100%|██████████| 22/22 [00:03<00:00,  6.46it/s]
100%|██████████| 22/22 [00:03<00:00,  6.44it/s]

Run time and cost

This model costs approximately $0.11 to run on Replicate, or 9 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 108 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion

Technion - Israel Institute of Technology

Abstract

Editing signals using large pre-trained models, in a zero-shot manner, has recently seen rapid advancements in the image domain. However, this wave has yet to reach the audio domain. In this paper, we explore two zero-shot editing techniques for audio signals, which use DDPM inversion on pre-trained diffusion models. The first, adopted from the image domain, allows text-based editing. The second, is a novel approach for discovering semantically meaningful editing directions without supervision. When applied to music signals, this method exposes a range of musically interesting modifications, from controlling the participation of specific instruments to improvisations on the melody.

Note: For now use input audio wav files