diff --git a/tutorials/intro.ipynb b/tutorials/intro.ipynb index d14308bb1..30dff8277 100644 --- a/tutorials/intro.ipynb +++ b/tutorials/intro.ipynb @@ -3,28 +3,34 @@ { "cell_type": "markdown", "metadata": { - "colab_type": "text", "id": "view-in-github" }, "source": [ "\"Open" ] }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZOWusGNiFdLN" + }, + "source": [ + "# Introduction to `pyannote.audio`" + ] + }, { "cell_type": "markdown", "metadata": { "id": "1Fs2d8otYnp7" }, "source": [ - "[`pyannote.audio`](https://github.com/pyannote/pyannote-audio) is an open-source toolkit written in Python for **speaker diarization**.\n", + "[`pyannote.audio`](https://github.com/pyannote/pyannote-audio) is an open-source Python toolkit for **speaker diarization** — the task of determining *“who speaks when”* by partitioning an audio conversation into speaker-specific time segments. \n", "\n", - "Based on [`PyTorch`](https://pytorch.org) machine learning framework, it provides a set of trainable end-to-end neural building blocks that can be combined and jointly optimized to build speaker diarization pipelines.\n", + "Based on the [`PyTorch`](https://pytorch.org) machine learning framework, it offers a collection of trainable, end-to-end neural building blocks. These components can be combined and jointly optimized to create powerful speaker diarization pipelines. \n", "\n", - "`pyannote.audio` also comes with pretrained [models](https://huggingface.co/models?other=pyannote-audio-model) and [pipelines](https://huggingface.co/models?other=pyannote-audio-pipeline) covering a wide range of domains for voice activity detection, speaker segmentation, overlapped speech detection, speaker embedding reaching state-of-the-art performance for most of them.\n", + "In addition, `pyannote.audio` provides pretrained [models](https://huggingface.co/models?other=pyannote-audio-model) and [pipelines](https://huggingface.co/models?other=pyannote-audio-pipeline) for a wide range of tasks such as voice activity detection, speaker segmentation, overlapped speech detection, and speaker embedding — many of which achieve state-of-the-art performance. \n", "\n", - "**This notebook will teach you how to apply those pretrained pipelines on your own data.**\n", - "\n", - "Make sure you run it using a GPU (or it might otherwise be slow...)" + "**This notebook will show you how to apply these pretrained pipelines to your own audio data.**\n" ] }, { @@ -44,137 +50,120 @@ }, "outputs": [], "source": [ - "!pip install -qq pyannote.audio==3.1.1\n", + "# install pyannote.audio 4.0\n", + "!pip install -qq git+https://github.com/pyannote/pyannote-audio.git@develop\n", + "\n", + "# install ipyannote, an interactive visualization tool for pyannote\n", + "!pip install -qq ipyannote\n", + "\n", "!pip install -qq ipython==7.34.0" ] }, { "cell_type": "markdown", "metadata": { - "id": "qggK-7VBYnp8" + "id": "qIzuCWUzFPHU" }, "source": [ - "# Visualization with `pyannote.core`\n", + "**⚠️ If you are running this notebook on Colab, restart the session (Runtime > Restart session), to avoid any dependencies errors in the rest of this tutorial.**" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5MclWK2GYnp_" + }, + "source": [ + "## Hugging Face setup\n", + "\n", + "Official [pyannote.audio](https://github.com/pyannote/pyannote-audio) pipelines (i.e. those under the [`pyannote` organization](https://hf.co/pyannote) umbrella) are open-source, but gated. It means that you have to first accept users conditions on their respective Hugging Face page to access the pretrained weights and hyper-parameters.\n", + "\n", + "For instance, to load the speaker diarization pipelines used in this tutorial, you have to visit [hf.co/pyannote/speaker-diarization-community-1](https://hf.co/pyannote/speaker-diarization-community-1) and accept the terms. Do the same for [hf.co/pyannote/speaker-diarization-community-1-cloud](https://hf.co/pyannote/speaker-diarization-community-1-cloud) and [hf.co/pyannote/speaker-diarization-precision-2](https://hf.co/pyannote/speaker-diarization-precision-2).\n", "\n", - "For the purpose of this notebook, we will download and use an audio file coming from the [AMI corpus](http://groups.inf.ed.ac.uk/ami/corpus/), which contains a conversation between 4 people in a meeting room." + "Finally log in using `notebook_login` below:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "id": "uJWoQiJgYnp8" + "id": "r5u7VMb-YnqB" }, "outputs": [], "source": [ - "!wget -q http://groups.inf.ed.ac.uk/ami/AMICorpusMirror/amicorpus/ES2004a/audio/ES2004a.Mix-Headset.wav\n", - "DEMO_FILE = {'uri': 'ES2004a.Mix-Headset', 'audio': 'ES2004a.Mix-Headset.wav'}" + "from huggingface_hub import notebook_login\n", + "notebook_login()" ] }, { "cell_type": "markdown", "metadata": { - "id": "EPIapoCJYnp8" + "id": "qggK-7VBYnp8" }, "source": [ - "Because AMI is a benchmarking dataset, it comes with manual annotations (a.k.a *groundtruth*). \n", - "Let us load and visualize the expected output of the speaker diarization pipeline.\n" + "## How to use `pyannote/speaker-diarization-community-1` ?" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": { - "id": "Mmm0Q22JYnp8" + "id": "xtZXNtEtOnlZ" }, - "outputs": [], "source": [ - "!wget -q https://raw.githubusercontent.com/pyannote/AMI-diarization-setup/main/only_words/rttms/test/ES2004a.rttm" + "Firstly, load the pipeline:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 233 - }, - "id": "ToqCwl_FYnp9", - "outputId": "a1d9631f-b198-44d1-ff6d-ec304125a9f4" + "id": "y5w-IGE1Ov25" }, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "execution_count": 10, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ - "# load groundtruth\n", - "from pyannote.database.util import load_rttm\n", - "_, groundtruth = load_rttm('ES2004a.rttm').popitem()\n", + "from pyannote.audio import Pipeline\n", + "import torch\n", "\n", - "# visualize groundtruth\n", - "groundtruth" + "pipeline = Pipeline.from_pretrained(\"pyannote/speaker-diarization-community-1\", skip_dependencies=True)\n", + "\n", + "# send pipeline to GPU (when available)\n", + "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "pipeline.to(torch.device(device))" ] }, { "cell_type": "markdown", "metadata": { - "id": "p_R9T9Y5Ynp9" + "id": "3ctcF5fVQFdg" }, "source": [ - "For the rest of this notebook, we will only listen to and visualize a one-minute long excerpt of the file (but will process the whole file anyway)." + "Apply the pipeline on an audio file:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 230 - }, - "id": "bAHza4Y1Ynp-", - "outputId": "c4cc2369-bfe4-4ac2-bb71-37602e7c7a8a" + "id": "xyn8ufT1QNIr" }, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "execution_count": 4, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ - "from pyannote.core import Segment, notebook\n", - "# make notebook visualization zoom on 600s < t < 660s time range\n", - "EXCERPT = Segment(600, 660)\n", - "notebook.crop = EXCERPT\n", + "from pyannote.audio.sample import SAMPLE_FILE\n", "\n", - "# visualize excerpt groundtruth\n", - "groundtruth" + "audio = SAMPLE_FILE[\"audio\"]\n", + "\n", + "# check https://github.com/pyannote/pyannote-audio/blob/853b2ab42c3ccd9ec898459d0ad24adc65167b3d/pyannote/audio/core/io.py#L45\n", + "# to see all accepted input types.\n", + "outputs = pipeline(audio)" ] }, { "cell_type": "markdown", "metadata": { - "id": "L3FQXT5FYnp-" + "id": "a6-rL-_aU0_Z" }, "source": [ - "This nice visualization is brought to you by [`pyannote.core`](http://pyannote.github.io/pyannote-core/) and basically indicates when each speaker speaks." + "We can then visualize `outputs.speaker_diarization`:" ] }, { @@ -183,399 +172,231 @@ "metadata": { "colab": { "base_uri": "https://localhost:8080/", - "height": 62 + "height": 136 }, - "id": "rDhZ3bXEYnp-", - "outputId": "a82efe4e-2f9c-48bd-94fb-c62af3a3cb43" + "id": "OV3sOnhLWULo", + "outputId": "94479f85-b5cb-43e6-ca89-690e68016b3e" }, "outputs": [ { "data": { - "text/html": [ - "\n", - " \n", - " " - ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAABiIAAADyCAYAAADAzN2uAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAHRhJREFUeJzt3XuQlfV9P/D34SphL8iS3WUFFS9RSUCNbZVcjDEKKuMEpVqT0Ui8ZMIgHTGNTByiMTamZWpsU7XmgpfGaPJzEk1janOxYElEjWkt1aQ0bmnUIoti9iLKJcv+/nDYuAXhrLvPHpZ9vWbOzO7zfM/z/Zwz3/M9z3ne5zxPqaurqysAAAAAAAAFGFbpAgAAAAAAgH2XIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACiMIAIAAAAAACjMPh9EvPjii5k/f34OPPDAjB49Oo2NjZk1a1Z+9rOfJUkOPvjglEqllEqljB07Nu9+97tz7733dt//c5/7XPf6N96OPPLInfq65557Mnz48CxYsGCndStWrEipVEpra2v3snXr1mXatGk58cQT09bW1t1mV7f169fvVM/w4cMzefLkfOITn8jLL79c9nOyefPmLFiwIHV1damqqsrcuXPT0tLSo82zzz6b2bNn521ve1vq6+vz6U9/Or/73e/K7mOoMc52Vs44+9M//dMcd9xxGT16dI455piytw0AAAAADB4j+rqBzo0b+6OOsgyvq+v1febOnZutW7fmzjvvzCGHHJKWlpY89NBD2fiGuj//+c/n0ksvTXt7e2644Yb8yZ/8SQ444IC85z3vSZK8853vzE9+8pMe2x0xYuenbtmyZbnyyivzla98JTfccEP222+/N62rubk5p556aqZOnZp77703Y8aM6V63Zs2a1NTU9GhfX1/f/feOejo7O/OrX/0qF110Udra2vLtb3+7rOdk0aJF+cEPfpB77703tbW1ueyyy3L22Wd3HzTv7OzM7Nmz09jYmEceeSQvvPBCPvaxj2XkyJG5/vrry+qjP/1209YB7W//saN6fR/jbGd7Gmc7XHTRRXnssceyevXqsrYLAAAAAAwufQ4i1k8/ph/KKM8B//tcr9q3trZm5cqVWbFiRT7wgQ8kSQ466KD80R/9UY921dXVaWxsTGNjY26++ebcdddd+f73v999gHjEiBFpbGzcbV9r167NI488ku985ztZvnx5vvvd7+ajH/3oLtuuXr06s2bNysknn5w777xzp4PN9fX1GTdu3Jv29cZ6DjjggJxzzjm5/fbbd1vfDm1tbVm2bFnuvvvunHzyyUmS22+/PUcddVQeffTRnHDCCfnRj36UX/7yl/nJT36ShoaGHHPMMbnuuuuyePHifO5zn8uoUb0/UN8Xpy9dPqD9PXrtrF61N852Vs44S5Ivf/nLSV7/RYkgAgAAAAD2Tfv0qZmqqqpSVVWV+++/P1u2bCnrPiNGjMjIkSOzdWvvvoV/++23Z/bs2amtrc3555+fZcuW7bLdI488kg984AOZO3du7rrrrl1+4703/ud//ic//OEPyw4HfvGLX2Tbtm055ZRTupcdeeSROfDAA7Nq1aokyapVqzJt2rQ0NDR0t5k1a1ba29vz9NNP96nefZFxtrNyxhkAAAAAMDTs00HEiBEjcscdd+TOO+/MuHHj8t73vjdXXXXVm37zeuvWrfniF7+Ytra27m9xJ8l//Md/dB9s3nH75Cc/2b1++/btueOOO3L++ecnSc4777z89Kc/zdq1a3fq46yzzsqZZ56Zm266KaVSaZd1TJo0qUdf73znO3us31HPmDFjMmXKlDz99NNZvHhxWc/J+vXrM2rUqJ2+Cd/Q0NB9fYD169f3CCF2rN+xjp6Ms52VM84AAAAAgKGhz6dm2tvNnTs3s2fPzsqVK/Poo4/mwQcfzNKlS/P1r3898+bNS5IsXrw4S5YsyebNm1NVVZW/+Iu/yOzZs7u3ccQRR+Qf/uEfemz3jefW//GPf5xNmzbljDPOSJJMmDAhp556am677bZcd911Pe734Q9/OPfdd19WrlyZ97///buseeXKlamuru7+f+TIkT3W76hn8+bNueuuu/Lkk09m4cKFvX9y6DfGGQAAAADArvU5iGhc/WQ/lFGs/fbbL6eeempOPfXUfPazn80ll1ySa665pvsA8ac//enMmzcvVVVVaWho2Okb5KNGjcphhx32pttftmxZXn755R4XAt6+fXtWr16da6+9NsOG/f6HJ1/5yldy5ZVX5vTTT88//uM/5sQTT9xpe1OmTNntufvfWM+Og9nXXnvtTgejd6WxsTFbt25Na2trjz5aWlq6rwfQ2NiYxx9/vMf9WlpautcNtAev/OCA9/lWGGe/V844AwAAAACGhj4HEcPr6vqjjgE1derU3H///d3/T5gwYbcHgHdn48aN+d73vpdvfetbPU5t09nZmfe973350Y9+lNNOO617ealUyle/+tUMGzYsZ5xxRn7wgx90X+D4rVqyZElOPvnkzJ8/P01NTbtte9xxx2XkyJF56KGHMnfu3CTJmjVr8uyzz2bGjBlJkhkzZuQLX/hCNmzYkPr6+iSvfxu/pqYmU6dO7VOtb8X+Ywf24tj9xTjb/TgDAAAAAIaGffrUTBs3bsw555yTiy66KNOnT091dXWeeOKJLF26NB/+8IfL3s7vfve7nc5rXyqV0tDQkG984xupq6vLueeeu9M33M8444wsW7asxwHiHfe99dZbM3z48O6DxCeddFL3+g0bNmTz5s097lNXV7fTqXN2mDFjRqZPn57rr78+N910024fS21tbS6++OJcccUVGT9+fGpqarJw4cLMmDEjJ5xwQpJk5syZmTp1ai644IIsXbo069evz5IlS7JgwYKMHj16t9sfioyznZUzzpLkmWeeySuvvJL169fntddey5NPPpnk9RCn3AtjAwAAAAB7t306iKiqqsrxxx+fG2+8Mc3Nzdm2bVsmT56cSy+9NFdddVXZ23n66aczceLEHstGjx6dzZs357bbbstZZ521ywsCz507NxdccEFeeumlndaVSqXcfPPNGTZsWGbPnp0HHnigextHHHHETu1XrVrV4wDu/7Vo0aLMmzcvixcvzuTJk3f7eG688cYMGzYsc+fOzZYtWzJr1qzccsst3euHDx+eBx54IPPnz8+MGTMyduzYXHjhhfn85z+/2+0OVcbZru1pnCXJJZdckocffrj7/2OPPTZJsnbt2hx88MG73T4AAAAAMDiUurq6uipdBAAAAAAAsG8atucmAAAAAAAAb40gYh/zzW9+M1VVVbu8vfEix9AXxhkAAAAAUC6nZtrHdHR0pKWlZZfrRo4cmYMOOmiAK2JfZJwBAAAAAOUSRAAAAAAAAIVxaiYAAAAAAKAwgggAAAAAAKAwI8pptH379qxbty7V1dUplUpF1wQAAAAAAOzFurq60tHRkaampgwbtvvfPJQVRKxbty6TJ0/ul+IAAAAAAIB9w3PPPZdJkybttk1ZQUR1dXX3BmtqavpeGQAAAAAAMGi1t7dn8uTJ3fnB7pQVROw4HVNNTY0gAgAAAAAASJKyLufgYtUAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhehVE3Hr936flv5/vc6cvNj+XFZctyYvNz+22XWdLS9pv+FI6W1r63Oee7Ohr3c9X56YvfKNfHudAK/d5pfJa/vv5QTvOivZSx5Z8bfkzealjS6VLgSHBaw4Gp3JeuwO5Lw30Xn+8Rr2PAzBUeQ8cfHoVRPy/NGXD833/IPPb3/xvDr/vzvz2N/+723adGzak40s3pnPDhj73uSc7+mpZ89+5a2t9vzzOgVbu80rlbXi+ZdCOs6K91LEly1Y0eyOBAeI1B4NTOa/dgdyXBnqvP16j3scBGKq8Bw4+Ts0EAAAAAAAURhABAAAAAAAURhABAAAAAAAUZkRv7/DK1u357aatfer01c2/y9gkpfa2dG7c+Kbttre29amft+S115KM6ZfHOdDKfV6pvK6OV5L0z+tpX9Px2rZKlwBDUsdr28xHMIj05v1ye6t9Q9gb9efnXe/jAAw1jh8NPr0OIhatfDlZubxPnU556Tf5qyRvW3BJ1vdpS/3vlb+9KTnr6n55nANtb35e6amt7sBBO86AfdPCv3+i0iUABdl43kcqXQJQMO/jAMDezqmZAAAAAACAwggiAAAAAACAwggiAAAAAACAwvT6GhE3vn98jpxxdJ86/c3Djyf3J6/e/PUc8v4/eNN22375qwE/p23VwsuS5/vncQ60cp9XKm/jo08lj28elOOsaM+s73COW6iAv/3YH+SwxupKlwGUqTfvl3Xfuicjpx5VcEVAb/Xn513v4wAMNY4fDT69DiKqRg3L/mNH9anTF/d7vduumtoMr6t703ad42r71M9bMmZMkv55nAOt3OeVyitVVyXZPCjHWdGqx4ysdAkwJFWPGWk+gkGkN++Xw8bZN4S9UX9+3vU+DsBQ4/jR4OPUTAAAAAAAQGEEEQAAAAAAQGEEEQAAAAAAQGEEEQAAAAAAQGF6FUScm3Wpn9TQ5073P+iA/PqsC7P/QQfstt3w+vpUX7Eow+vr+9znnuzoq+GIQ3L+qA398jgHWrnPK5VXP6lh0I6zok2oHp2LTzo0E6pHV7oUGBK85mBwKue1O5D70kDv9cdr1Ps4AEOV98DBp9TV1dW1p0bt7e2pra1NW1tbampqBqIuAAAAAABgL9Wb3MCpmQAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIgAAAAAAgMIIIoB0trSk/YYvpbOlpdKlAEOEeQf2Di91bMnXlj+Tlzq27FXbAgBg4PmcRm91bthQdltBBJDODRvS8aUbezV5APSFeQf2Di91bMmyFc39FkT017YAABh4PqfRW50vvlh2W0EEAAAAAABQGEEEAAAAAABQmBGVLgDYe2xvbUvnxo2VLgMYAra3tlW6BOANOl7blt9u2trnbQAAMPg5PkS5tre1l91WEAF023jeRypdAgBQAQv//olKlwAAwF7C8SHK1bF9e9ltnZoJAAAAAAAojCACAAAAAAAojCACAAAAAAAojGtEAN3qvnVPRk49qtJlAEPAtl/+ynlHYS/ytx/7gxzWWN2nbTyzvsO1JgAA9gGOD1GukT9/Ijn9tLLaCiKAbsPG1WZ4XV2lywCGgM5xtZUuAXiD6jEjs//YUX3eBgAAg5/jQ5RrWG1N+W0LrAMAAAAAABjiBBEAAAAAAEBhBBEAAAAAAEBhBBEAAAAAAEBhBBFAhtfXp/qKRRleX1/pUoAhwrwDe4cJ1aNz8UmHZkL16L1qWwAADDyf0+it4W9/e9ltS11dXV17atTe3p7a2tq0tbWlpqb8K2EDAAAAAAD7nt7kBn4RAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFEYQAQAAAAAAFGZEOY26urqSJO3t7YUWAwAAAAAA7P125AU78oPdKSuI6OjoSJJMnjy5D2UBAAAAAAD7ko6OjtTW1u62TamrjLhi+/btWbduXaqrq1MqlfqtQOD32tvbM3ny5Dz33HOpqampdDkAez3zJkD5zJkAvWPeBNizrq6udHR0pKmpKcOG7f4qEGX9ImLYsGGZNGlSvxQH7F5NTY2dHIBeMG8ClM+cCdA75k2A3dvTLyF2cLFqAAAAAACgMIIIAAAAAACgMIII2EuMHj0611xzTUaPHl3pUgAGBfMmQPnMmQC9Y94E6F9lXawaAAAAAADgrfCLCAAAAAAAoDCCCAAAAAAAoDCCCAAAAAAAoDCCCAAAAAAAoDCCCBhg//Iv/5IzzzwzTU1NKZVKuf/++3us7+rqytVXX52JEydmzJgxOeWUU/LrX/+6MsUCVNie5sx58+alVCr1uJ122mmVKRagwr74xS/mD//wD1NdXZ36+vrMmTMna9as6dFm8+bNWbBgQerq6lJVVZW5c+empaWlQhUDVFY58+ZJJ5200/7mJz/5yQpVDDB4CSJggG3atClHH310br755l2uX7p0ab785S/n1ltvzWOPPZaxY8dm1qxZ2bx58wBXClB5e5ozk+S0007LCy+80H275557BrBCgL3Hww8/nAULFuTRRx/Nj3/842zbti0zZ87Mpk2butssWrQo3//+93Pvvffm4Ycfzrp163L22WdXsGqAyiln3kySSy+9tMf+5tKlSytUMcDgVerq6uqqdBEwVJVKpdx3332ZM2dOktd/DdHU1JRPfepT+bM/+7MkSVtbWxoaGnLHHXfkvPPOq2C1AJX1f+fM5PVfRLS2tu70SwkAkhdffDH19fV5+OGHc+KJJ6atrS1vf/vbc/fdd+eP//iPkyT/+Z//maOOOiqrVq3KCSecUOGKASrr/86byeu/iDjmmGPy13/915UtDmCQ84sI2IusXbs269evzymnnNK9rLa2Nscff3xWrVpVwcoA9l4rVqxIfX19jjjiiMyfPz8bN26sdEkAe4W2trYkyfjx45Mkv/jFL7Jt27Ye+5pHHnlkDjzwQPuaANl53tzhm9/8ZiZMmJB3vetd+cxnPpNXX321EuUBDGojKl0A8Hvr169PkjQ0NPRY3tDQ0L0OgN877bTTcvbZZ2fKlClpbm7OVVddldNPPz2rVq3K8OHDK10eQMVs3749l19+ed773vfmXe96V5LX9zVHjRqVcePG9WhrXxNg1/Nmknz0ox/NQQcdlKampqxevTqLFy/OmjVr8t3vfreC1QIMPoIIAGDQeuMp66ZNm5bp06fn0EMPzYoVK/KhD32ogpUBVNaCBQvy1FNP5ac//WmlSwEYFN5s3vzEJz7R/fe0adMyceLEfOhDH0pzc3MOPfTQgS4TYNByaibYizQ2NiZJWlpaeixvaWnpXgfAmzvkkEMyYcKEPPPMM5UuBaBiLrvssjzwwANZvnx5Jk2a1L28sbExW7duTWtra4/29jWBoe7N5s1dOf7445PE/iZALwkiYC8yZcqUNDY25qGHHupe1t7ensceeywzZsyoYGUAg8Pzzz+fjRs3ZuLEiZUuBWDAdXV15bLLLst9992Xf/7nf86UKVN6rD/uuOMycuTIHvuaa9asybPPPmtfExiS9jRv7sqTTz6ZJPY3AXrJqZlggL3yyis9vjmxdu3aPPnkkxk/fnwOPPDAXH755fnzP//zHH744ZkyZUo++9nPpqmpKXPmzKlc0QAVsrs5c/z48bn22mszd+7cNDY2prm5OVdeeWUOO+ywzJo1q4JVA1TGggULcvfdd+d73/teqquru6/7UFtbmzFjxqS2tjYXX3xxrrjiiowfPz41NTVZuHBhZsyYkRNOOKHC1QMMvD3Nm83Nzbn77rtzxhlnpK6uLqtXr86iRYty4oknZvr06RWuHmBwKXV1dXVVuggYSlasWJEPfvCDOy2/8MILc8cdd6SrqyvXXHNNvvrVr6a1tTXve9/7csstt+Qd73hHBaoFqKzdzZl/93d/lzlz5uTf/u3f0tramqampsycOTPXXXddGhoaKlAtQGWVSqVdLr/99tszb968JMnmzZvzqU99Kvfcc0+2bNmSWbNm5ZZbbnFqJmBI2tO8+dxzz+X888/PU089lU2bNmXy5Mk566yzsmTJktTU1AxwtQCDmyACAAAAAAAojGtEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAAAAAhRFEAAAAPcybNy9z5sypdBkAAMA+YkSlCwAAAAZOqVTa7fprrrkmf/M3f5Ourq4BqggAANjXCSIAAGAIeeGFF7r//va3v52rr746a9as6V5WVVWVqqqqSpQGAADso5yaCQAAhpDGxsbuW21tbUqlUo9lVVVVO52a6aSTTsrChQtz+eWXZ//9909DQ0O+9rWvZdOmTfn4xz+e6urqHHbYYXnwwQd79PXUU0/l9NNPT1VVVRoaGnLBBRfkpZdeGuBHDAAAVJogAgAA2KM777wzEyZMyOOPP56FCxdm/vz5Oeecc/Ke97wn//qv/5qZM2fmggsuyKuvvpokaW1tzcknn5xjjz02TzzxRP7pn/4pLS0tOffccyv8SAAAgIEmiAAAAPbo6KOPzpIlS3L44YfnM5/5TPbbb79MmDAhl156aQ4//PBcffXV2bhxY1avXp0kuemmm3Lsscfm+uuvz5FHHpljjz02t912W5YvX57/+q//qvCjAQAABpJrRAAAAHs0ffr07r+HDx+eurq6TJs2rXtZQ0NDkmTDhg1Jkn//93/P8uXLd3m9iebm5rzjHe8ouGIAAGBvIYgAAAD2aOTIkT3+L5VKPZaVSqUkyfbt25Mkr7zySs4888z85V/+5U7bmjhxYoGVAgAAextBBAAA0O/e/e535zvf+U4OPvjgjBjhYwcAAAxlrhEBAAD0uwULFuTll1/ORz7ykfz85z9Pc3NzfvjDH+bjH/94Ojs7K10eAAAwgAQRAABAv2tqasrPfvazdHZ2ZubMmZk2bVouv/zyjBs3LsOG+RgCAABDSamrq6ur0kUAAAAAAAD7Jl9FAgAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACiOIAAAAAAAACvP/AUKEbE6xij1uAAAAAElFTkSuQmCC", "text/plain": [ - "" + "" ] }, - "execution_count": 11, + "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "from pyannote.audio import Audio\n", - "from IPython.display import Audio as IPythonAudio\n", - "waveform, sr = Audio(mono=\"downmix\").crop(DEMO_FILE, EXCERPT)\n", - "IPythonAudio(waveform.flatten(), rate=sr)" + "outputs.speaker_diarization" ] }, { "cell_type": "markdown", "metadata": { - "id": "hkzox7QIYnp_" + "id": "9qQct_dbTd21" }, "source": [ - "# Processing your own audio file (optional)\n", - "\n", - "In case you just want to go ahead with the demo file, skip this section entirely.\n", + "Or visualize the output in a more interactive way using `ipyannote` widget:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "N_XryMONTcue" + }, + "outputs": [], + "source": [ + "from ipyannote import IPyannote\n", "\n", - "In case you want to try processing your own audio file, proceed with running this section. It will offer you to upload an audio file (preferably a `wav` file but all formats supported by [`SoundFile`](https://pysoundfile.readthedocs.io/en/latest/) should work just fine)." + "IPyannote(audio=audio, annotation=outputs.speaker_diarization)" ] }, { "cell_type": "markdown", "metadata": { - "id": "3hmFmLzFYnp_" + "id": "_ajQeskZU5Y0" }, "source": [ - "## Upload audio file" + "And even compare with the reference using `ipyannote.Errors`:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "id": "xC05jFO_Ynp_" + "id": "Ku7Q6pP4VCbP" }, "outputs": [], "source": [ - "import google.colab\n", - "own_file, _ = google.colab.files.upload().popitem()\n", - "OWN_FILE = {'audio': own_file}\n", - "notebook.reset()\n", + "from ipyannote import Errors\n", "\n", - "# load audio waveform and play it\n", - "waveform, sample_rate = Audio(mono=\"downmix\")(OWN_FILE)\n", - "IPythonAudio(data=waveform.squeeze(), rate=sample_rate, autoplay=True)" + "reference = SAMPLE_FILE[\"annotation\"]\n", + "\n", + "Errors(audio=audio, reference=reference.rename_tracks(), hypothesis=outputs.speaker_diarization)" ] }, { "cell_type": "markdown", "metadata": { - "id": "ctw4nLaPYnp_" + "id": "q5KSipqrYiTU" }, "source": [ - "Simply replace `DEMO_FILE` by `OWN_FILE` in the rest of the notebook.\n", - "\n", - "Note, however, that unless you provide a groundtruth annotation in the next cell, you will (obviously) not be able to visualize groundtruth annotation nor evaluate the performance of the diarization pipeline quantitatively" + "In the visualizer above, the first line shows the reference, the second one displays the pipeline output, and the last line highlights the errors made by the pipeline compared to the reference." ] }, { "cell_type": "markdown", "metadata": { - "id": "x9AQgDzFYnp_" + "id": "P0dB68V3bZsO" }, "source": [ - "## Upload groundtruth (optional)\n", - "\n", - "The groundtruth file is expected to use the RTTM format, with one line per speech turn with the following convention:\n", - "\n", - "```\n", - "SPEAKER {file_name} 1 {start_time} {duration} {speaker_name} \n", - "```" + "You might want to get speaker diarization adapted to downstream transcription that does not contains any overlapping speech turns. This can be achieved using `outputs.exclusive_speaker_diarization`:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "id": "iZaFudpDYnp_", - "outputId": "981274fa-e654-4091-c838-91c81f921e5d" + "id": "ZkkYB14uZfcv" }, - "outputs": [ - { - "data": { - "text/html": [ - "\n", - " \n", - " \n", - " Upload widget is only available when the cell has been executed in the\n", - " current browser session. Please rerun this cell to enable.\n", - " \n", - " " - ], - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Saving sample.rttm to sample.rttm\n" - ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "execution_count": null, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ - "groundtruth_rttm, _ = google.colab.files.upload().popitem()\n", - "groundtruths = load_rttm(groundtruth_rttm)\n", - "if OWN_FILE['audio'] in groundtruths:\n", - " groundtruth = groundtruths[OWN_FILE['audio']]\n", - "else:\n", - " _, groundtruth = groundtruths.popitem()\n", - "groundtruth" + "IPyannote(audio=audio, annotation=outputs.exclusive_speaker_diarization)" ] }, { "cell_type": "markdown", "metadata": { - "id": "5MclWK2GYnp_" + "id": "3hmFmLzFYnp_" }, "source": [ - "# Speaker diarization with `pyannote.pipeline`\n", - "\n", - "We are about to run a full speaker diarization pipeline, that includes speaker segmentation, speaker embedding, and a final clustering step. **Brace yourself!**\n", - "\n", - "To load the speaker diarization pipeline,\n", - "\n", - "* accept the user conditions on [hf.co/pyannote/speaker-diarization-3.1](https://hf.co/pyannote/speaker-diarization-3.1)\n", - "* accept the user conditions on [hf.co/pyannote/segmentation-3.0](https://hf.co/pyannote/segmentation-3.0)\n", - "* login using `notebook_login` below" + "## A word about hosted `speaker-diarization-community-1-cloud`" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 301, - "referenced_widgets": [ - "c8731777ce834e58a76a295076200cfc", - "859b12a6d95b4c6f987791ca848122b9", - "94756148d2e94a93ae233baba20af683", - "ba18cded436e486da34882d821d8f1eb", - "99898e6ee64a46bd832af112e79b58b7", - "79184c8c2a6f4b7493bb7f6983f18a09", - "ea95ffd922c0455d957120f034e541f8", - "13525aa369a9410a83343952ab511f3c", - "b2be65e192384c948fb8987d4cfca505", - "333b42ca7aa44788b1c22724eb11bcc3", - "0e382d66f09f4958a40baa7ab83c4ccb", - "6a45ce374e2e47ba9457d02e02522748", - "765485a1d3f941d28b79782dcffbf401", - "3499ef4dd9f243d9bef00b396e78ed69", - "6e56329c30c0441c8d45df3975e75a76" - ] - }, - "id": "r5u7VMb-YnqB", - "outputId": "c714a997-d4f8-417a-e5ad-0a4924333859" + "id": "t2Cz6_-7dipH" }, - "outputs": [ - { - "data": { - "application/vnd.jupyter.widget-view+json": { - "model_id": "6e56329c30c0441c8d45df3975e75a76", - "version_major": 2, - "version_minor": 0 - }, - "text/plain": [ - "VBox(children=(HTML(value='
" - ] - }, - "execution_count": 14, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ - "diarization" + "import os\n", + "\n", + "token = os.environ[\"PYANNOTEAI_API_KEY\"] # or simply paste your API key here" ] }, { "cell_type": "markdown", "metadata": { - "id": "DLhErS6wYnqB" + "id": "oGNYZJ1AgZOe" }, "source": [ - "# Evaluation with `pyannote.metrics`\n", - "\n", - "Because groundtruth is available, we can evaluate the quality of the diarization pipeline by computing the [diarization error rate](http://pyannote.github.io/pyannote-metrics/reference.html#diarization)." + "Then you can load the pipeline and process your audio:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "id": "vNHQRTUIYnqB" + "id": "xC05jFO_Ynp_" }, "outputs": [], "source": [ - "from pyannote.metrics.diarization import DiarizationErrorRate\n", - "metric = DiarizationErrorRate()\n", - "der = metric(groundtruth, diarization)" + "import os\n", + "cloud_pipeline = Pipeline.from_pretrained(\"pyannote/speaker-diarization-community-1\", token=token, skip_dependencies=True)\n", + "cloud_outputs = cloud_pipeline(audio)\n" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "9d0vKQ0fYnqB", - "outputId": "9a664753-cd84-4211-9153-d33e929bb252" + "id": "fTIH0rqWmowD" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "diarization error rate = 19.8%\n" - ] - } - ], "source": [ - "print(f'diarization error rate = {100 * der:.1f}%')" + "Visualize the speaker diarization outputs in the same way as before:" ] }, { - "cell_type": "markdown", + "cell_type": "code", + "execution_count": null, "metadata": { - "id": "Xz5QJV9nYnqB" + "id": "6h9JVnrOmubd" }, + "outputs": [], "source": [ - "This implementation of diarization error rate is brought to you by [`pyannote.metrics`](http://pyannote.github.io/pyannote-metrics/).\n", - "\n", - "It can also be used to improve visualization by find the optimal one-to-one mapping between groundtruth and hypothesized speakers." + "IPyannote(audio=audio, annotation=cloud_outputs.speaker_diarization)" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 230 - }, - "id": "xMLf4mrYYnqB", - "outputId": "ed08bcc8-24c6-439c-a244-3a673ff480b0" + "id": "WIBcxzi1m7kV" }, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "execution_count": 19, - "metadata": {}, - "output_type": "execute_result" - } - ], "source": [ - "mapping = metric.optimal_mapping(groundtruth, diarization)\n", - "diarization.rename_labels(mapping=mapping)" + "Want something even more precise? Try the `pyannote/speaker-diarization-precision-2` pipeline:\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 230 - }, - "id": "Z0ewsLlQYnqB", - "outputId": "8a8cd040-ee1d-48f7-d4be-eef9e08e9e55" + "id": "rtH_51m4nXAJ" }, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "execution_count": 20, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ - "groundtruth" + "precision_pipeline = Pipeline.from_pretrained(\"pyannote/speaker-diarization-precision-2\", token=token, skip_dependencies=True)\n", + "precision_outputs = precision_pipeline(audio)\n", + "\n", + "IPyannote(audio=audio, annotation=precision_outputs.speaker_diarization)" ] }, { @@ -584,22 +405,23 @@ "id": "MxlrTbyPYnqB" }, "source": [ - "# Going further\n", + "## Going further\n", "\n", "We have only scratched the surface in this introduction.\n", "\n", - "More details can be found in the [`pyannote.audio` Github repository](https://github.com/pyannote/pyannote-audio).\n" + "More details can be found in the [`pyannote.audio` Github repository](https://github.com/pyannote/pyannote-audio).\n", + "\n", + "You can also visit the [`pyannoteAI`](https://www.pyannote.ai/) website to explore our fastest and most advanced solutions.\n" ] } ], "metadata": { "accelerator": "GPU", "colab": { - "include_colab_link": true, "provenance": [] }, "kernelspec": { - "display_name": "Python 3 (ipykernel)", + "display_name": "pyannote-audio", "language": "python", "name": "python3" }, @@ -613,442 +435,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.13" - }, - "widgets": { - "application/vnd.jupyter.widget-state+json": { - "0e382d66f09f4958a40baa7ab83c4ccb": { - "model_module": "@jupyter-widgets/base", - "model_module_version": "1.2.0", - "model_name": "LayoutModel", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "13525aa369a9410a83343952ab511f3c": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "DescriptionStyleModel", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "333b42ca7aa44788b1c22724eb11bcc3": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "DescriptionStyleModel", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "3499ef4dd9f243d9bef00b396e78ed69": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "DescriptionStyleModel", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "6a45ce374e2e47ba9457d02e02522748": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "ButtonStyleModel", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ButtonStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "button_color": null, - "font_weight": "" - } - }, - "765485a1d3f941d28b79782dcffbf401": { - "model_module": "@jupyter-widgets/base", - "model_module_version": "1.2.0", - "model_name": "LayoutModel", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "79184c8c2a6f4b7493bb7f6983f18a09": { - "model_module": "@jupyter-widgets/base", - "model_module_version": "1.2.0", - "model_name": "LayoutModel", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": "center", - "align_self": null, - "border": null, - "bottom": null, - "display": "flex", - "flex": null, - "flex_flow": "column", - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": "50%" - } - }, - "859b12a6d95b4c6f987791ca848122b9": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "HTMLModel", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ea95ffd922c0455d957120f034e541f8", - "placeholder": "​", - "style": "IPY_MODEL_13525aa369a9410a83343952ab511f3c", - "value": "

Copy a token from your Hugging Face\ntokens page and paste it below.
Immediately click login after copying\nyour token or it might be stored in plain text in this notebook file.
" - } - }, - "94756148d2e94a93ae233baba20af683": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "PasswordModel", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "PasswordModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "PasswordView", - "continuous_update": true, - "description": "Token:", - "description_tooltip": null, - "disabled": false, - "layout": "IPY_MODEL_b2be65e192384c948fb8987d4cfca505", - "placeholder": "​", - "style": "IPY_MODEL_333b42ca7aa44788b1c22724eb11bcc3", - "value": "" - } - }, - "99898e6ee64a46bd832af112e79b58b7": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "HTMLModel", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_765485a1d3f941d28b79782dcffbf401", - "placeholder": "​", - "style": "IPY_MODEL_3499ef4dd9f243d9bef00b396e78ed69", - "value": "\nPro Tip: If you don't already have one, you can create a dedicated\n'notebooks' token with 'write' access, that you can then easily reuse for all\nnotebooks.
" - } - }, - "b2be65e192384c948fb8987d4cfca505": { - "model_module": "@jupyter-widgets/base", - "model_module_version": "1.2.0", - "model_name": "LayoutModel", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "ba18cded436e486da34882d821d8f1eb": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "ButtonModel", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ButtonModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ButtonView", - "button_style": "", - "description": "Login", - "disabled": false, - "icon": "", - "layout": "IPY_MODEL_0e382d66f09f4958a40baa7ab83c4ccb", - "style": "IPY_MODEL_6a45ce374e2e47ba9457d02e02522748", - "tooltip": "" - } - }, - "c8731777ce834e58a76a295076200cfc": { - "model_module": "@jupyter-widgets/controls", - "model_module_version": "1.5.0", - "model_name": "VBoxModel", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "VBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "VBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_859b12a6d95b4c6f987791ca848122b9", - "IPY_MODEL_94756148d2e94a93ae233baba20af683", - "IPY_MODEL_ba18cded436e486da34882d821d8f1eb", - "IPY_MODEL_99898e6ee64a46bd832af112e79b58b7" - ], - "layout": "IPY_MODEL_79184c8c2a6f4b7493bb7f6983f18a09" - } - }, - "ea95ffd922c0455d957120f034e541f8": { - "model_module": "@jupyter-widgets/base", - "model_module_version": "1.2.0", - "model_name": "LayoutModel", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - } - } + "version": "3.10.18" } }, "nbformat": 4,