StableAudioSmall.ipynb

stableaudiosmall.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyMIyooOkbNOd0KvleXUJUsk",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/hathibelagal-dev/2baeb25cb26f474f6da46db248da2bfd/stableaudiosmall.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install torchsde"
      ],
      "metadata": {
        "id": "7uJmgn7sRNJz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import soundfile as sf\n",
        "from diffusers import StableAudioPipeline\n",
        "\n",
        "pipe = StableAudioPipeline.from_pretrained(\"stabilityai/stable-audio-open-1.0\", torch_dtype=torch.float16)"
      ],
      "metadata": {
        "id": "Nt2d_T-bQaVq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pipe.to(\"cuda\")"
      ],
      "metadata": {
        "id": "PtnO4f2zTGg7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "prompt = \"128 BPM tech house drum loop\"\n",
        "negative_prompt = \"Low quality.\""
      ],
      "metadata": {
        "id": "nw7iXhBDQm6U"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "audio = pipe(\n",
        "    prompt,\n",
        "    negative_prompt=negative_prompt,\n",
        "    num_inference_steps=50,\n",
        "    audio_end_in_s=10.0,\n",
        "    num_waveforms_per_prompt=3,\n",
        ").audios[0]"
      ],
      "metadata": {
        "id": "uSz87b8LQvDu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "output = audio.T.float().cpu().numpy()\n",
        "sf.write(\"beat.wav\", output, pipe.vae.sampling_rate)"
      ],
      "metadata": {
        "id": "iAVMsLo9Q1EY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from IPython.display import display, Audio\n",
        "display(Audio(\"beat.wav\"))"
      ],
      "metadata": {
        "id": "x-G1O84sTZS_"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}