---
language:
- en
library_name: stable-audio-tools
license: other
license_name: stable-audio-community
pipeline_tag: text-to-audio
tags:
  - text-to-audio
inference: true
widget:
- src: ./assets/demo_cfg_3_00000001.wav
  example_title: 'Unconditional (blank prompt)'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
- text: 'Chill soft wake up, slow down alt, night get lucky dance, relax music introspective 2017 2018 2019 2020 2021 2022, acoustic atmosphere uplifting dreams, dreamy indie pop, electric trap, percussion, higher reverb, really intensity melody, goodbye'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/music_3_illustration.jpg
- text: 'Chill hip-hop beat, chillhop, lofi pop, favorite music'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/music_4_illustration.jpg
---



<style>
.spoiler{background:black;color:black;text-decoration:none!important}.spoiler a{color:black;text-decoration:underline}.spoiler:hover,.spoiler:hover a{color:white}
</style>

You can use this model in [stable-audio-tools](https://github.com/Stability-AI/stable-audio-tools), fine-tuned on my favorite song <span class="spoiler">from my [personal playlist](https://www.youtube.com/watch?v=dQw4w9WgXcQ).</span>

<details>
  <summary>Comparison Table</summary>
  <table style="width:100%; border-collapse: collapse;">
    <colgroup>
      <col style="width: 25%;">
      <col style="width: 37.5%;">
      <col style="width: 37.5%;">
    </colgroup>
    <tr>
      <th>Prompt</th>
      <th>Base Model</th>
      <th>Fine-Tuned</th>
    </tr>
    <tr>
      <td style="font-size: smaller; padding: 0.5px; word-wrap: break-word;">
        Feel-Good Vibes and Dramatic Atmosphere, alone hero, epic, get good yeah, better last night pop, follow follow, echoing, powerful vocal driving melancholic vocals dramatic Features rising tension, progressive electro house, far away, by Alan Walker, popular song tempo, girl, female synth, popular, titled: legend never die
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/1_Base_stable-audio-open-1.0.wav" type="audio/wav">
        </audio>
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/1_1epoch.wav" type="audio/wav">
        </audio>
      </td>
    </tr>
    <tr>
      <td style="font-size: smaller; padding: 0.5px; word-wrap: break-word;">
        Beautiful music progressive electro slap mood, upbeat, heavy bass, melancholic, hopeful; drums, vocals, dynamic shifts, building intensity, run far away, repetitive, let let go, think of us, titled popular lyrics: Mirror's Edge, popular lyrics say: "still still alive"
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/2_Base_stable-audio-open-1.0.wav" type="audio/wav">
        </audio>
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/2_1epoch.wav" type="audio/wav">
        </audio>
      </td>
    </tr>
    <tr>
      <td style="font-size: smaller; padding: 0.5px; word-wrap: break-word;">
        Chill soft wake up, slow down alt, night get lucky dance, relax music introspective 2017 2018 2019 2020 2021 2022, acoustic atmosphere uplifting dreams, dreamy indie pop, electric trap, percussion, higher reverb, really intensity melody, goodbye
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/3_Base_stable-audio-open-1.0.wav" type="audio/wav">
        </audio>
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/3_1epoch.wav" type="audio/wav">
        </audio>
      </td>
    </tr>
    <tr>
      <td style="font-size: smaller; padding: 0.5px; word-wrap: break-word;">
        Chill hip-hop beat, chillhop, lofi pop, favorite music
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/4_Base_stable-audio-open-1.0.wav" type="audio/wav">
        </audio>
      </td>
      <td style="padding: 0.5px; vertical-align: middle; text-align: center;">
        <audio controls style="width: 100%;">
          <source src="https://huggingface.co/Nekochu/stable-audio-open-1.0-Music/resolve/main/assets/4_1epoch.wav" type="audio/wav">
        </audio>
      </td>
    </tr>
  </table>

<details open>
  <summary>Showcase Model Details</summary>
  <div>
    <h3>Test Settings:</h3>
    <ul>
      <li>CFG: 7.0</li>
      <li>Steps: 100</li>
      <li>Seed: -1</li>
    </ul>
    <p>Prompt have been chosen based on the top tagged words except last prompt which is used to compare effect on non-trained tags</p>
  </div>

  <div>
  <Gallery />
  <div class="not-prose mb-2 flex flex-wrap items-start gap-4 sm:mr-6 sm:flex-row">
    <audio controls style="width: calc(50% - 8px);">
      <source src="./assets/3_1epoch.wav" type="audio/wav">
    </audio>
    <audio controls style="width: calc(50% - 8px);">
      <source src="./assets/4_1epoch.wav" type="audio/wav">
    </audio>
  </div>
</div>
    
</details>
</details>

<details>
  <summary>Training</summary>

  ### Dataset: 2-3 min music length
  - All of my Liked music [download and auto label](https://pastebin.com/z1bkZyqe) so mostly copyright. 
  - Total number of samples: ~1383
  - `"random_crop": true` in [dataset_config.json](https://github.com/Stability-AI/stable-audio-tools/issues/99#issuecomment-2174885688)
  
  ### Settings:
  - Training epochs: 1
  - Training steps: 1383
  - Learning rate: 1e-05
    
</details>