make colab friendly

notebooks/test_mel.ipynb

@@ -1,14 +1,28 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "3de63898",
+   "metadata": {},
+   "source": [
+    "<a href=\"https://colab.research.google.com/github/teticio/audio-diffusion/blob/master/notebooks/test_mel.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "2a61d194",
+   "id": "81fbd495",
    "metadata": {},
    "outputs": [],
    "source": [
-    "%load_ext autoreload\n",
-    "%autoreload 2"
+    "try:\n",
+    "    # are we running on Google Colab?\n",
+    "    import google.colab\n",
+    "    !git clone -q https://github.com/teticio/audio-diffusion.git\n",
+    "    %cd audio-diffusion\n",
+    "    !pip install -q -r requirements.txt\n",
+    "except:\n",
+    "    pass"
    ]
   },
   {
@@ -30,6 +44,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "from datasets import load_dataset\n",
     "from IPython.display import Audio\n",
     "from audiodiffusion.mel import Mel"
    ]
@@ -58,6 +73,21 @@
     "### Transform slice of audio to mel spectrogram"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f32bb35e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "try:\n",
+    "    # are we running on Google Colab?\n",
+    "    from google.colab import files\n",
+    "    audio_file = list(files.upload().keys())[0]\n",
+    "except:\n",
+    "    audio_file = \"/home/teticio/Music/Music/A Tribe Called Quest/The Anthology/08 Can I Kick It_.mp3\""
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -65,7 +95,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "mel.load_audio('/home/teticio/Music/Music/A Tribe Called Quest/The Anthology/08 Can I Kick It_.mp3')"
+    "mel.load_audio(audio_file)"
    ]
   },
   {
@@ -108,10 +138,58 @@
     "Audio(data=audio, rate=mel.get_sample_rate())"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "0f1f2006",
+   "metadata": {},
+   "source": [
+    "### Select a random image from the training set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1f29f025",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ds = load_dataset('teticio/audio-diffusion-256')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e002482d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "image = ds['train'].shuffle().select(range(1))['image'][0]\n",
+    "image"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6a801fc5",
+   "metadata": {},
+   "source": [
+    "### Convert to audio"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a2421827",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "audio = mel.image_to_audio(image)\n",
+    "Audio(data=audio, rate=mel.get_sample_rate())"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "a0dffbc4",
+   "id": "2281fb55",
    "metadata": {},
    "outputs": [],
    "source": []