audio-course/pr_239/fr/chapter2/audio_classification_pipeline.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Classification audio avec un pipeline&quot;,&quot;local&quot;:&quot;classification-audio-avec-un-pipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}">
		<link href="/docs/audio-course/pr_239/fr/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/entry/start.48d22ca3.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/scheduler.a045fce0.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/singletons.932e46dd.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/index.2447c7d9.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/paths.50fd3c21.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/entry/app.4e1ef9b0.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/preload-helper.aa8f5fd5.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/index.5135b896.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/nodes/0.00a7cda2.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/nodes/13.ff1ca69b.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.59591d32.js">
		<link rel="modulepreload" href="/docs/audio-course/pr_239/fr/_app/immutable/chunks/CodeBlock.1d730892.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Classification audio avec un pipeline&quot;,&quot;local&quot;:&quot;classification-audio-avec-un-pipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END -->      <p></p>  <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg></button></div> </div>  <h1 class="relative group"><a id="classification-audio-avec-un-pipeline" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#classification-audio-avec-un-pipeline"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>Classification audio avec un pipeline</span></h1> <p data-svelte-h="svelte-rvzziw">La classification audio consiste à attribuer une ou plusieurs étiquettes à un enregistrement audio en fonction de son contenu.
Les étiquettes peuvent correspondre à différentes catégories sonores, telles que la musique, la parole ou le bruit, ou à des catégories plus spécifiques telles que le chant d’oiseaux ou les sons de moteur de voiture.
Avant de plonger dans les détails du fonctionnement des <em>transformers</em> audio les plus populaires, et avant de finetuner un modèle personnalisé, voyons comment vous pouvez utiliser un modèle pré-entraîné standard pour la classification audio avec seulement quelques lignes de code avec 🤗 <em>Transformers</em>.
Utilisons le même jeu de données <a href="https://huggingface.co/datasets/PolyAI/minds14" rel="nofollow">MINDS-14</a> que vous avez exploré dans l’unité précédente.
Si vous vous souvenez, MINDS-14 contient des enregistrements de personnes posant des questions à un système bancaire électronique dans plusieurs langues et dialectes, et a le <code>intent_class</code> pour chaque enregistrement. Nous pouvons classer les enregistrements par intention de l’appel.</p> <p data-svelte-h="svelte-wbkria">Comme précédemment, commençons par charger le sous-ensemble <code>en-AU</code> pour essayer le pipeline, et suréchantillonnons-le à un taux d’échantillonnage de 16 kHz, ce qui est ce que la plupart des modèles vocaux exigent.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset
<span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> Audio

minds = load_dataset(<span class="hljs-string">&quot;PolyAI/minds14&quot;</span>, name=<span class="hljs-string">&quot;en-AU&quot;</span>, split=<span class="hljs-string">&quot;train&quot;</span>)
minds = minds.cast_column(<span class="hljs-string">&quot;audio&quot;</span>, Audio(sampling_rate=<span class="hljs-number">16_000</span>))<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-12hj9jn">Pour classer un enregistrement audio dans un ensemble de classes, nous pouvons utiliser le pipeline <code>audio-classification</code> de 🤗 <em>Transformers</em>.
Dans notre cas, nous avons besoin d’un modèle qui a été finetuné pour la classification des intentions, et en particulier sur le jeu de données MINDS-14. Heureusement pour nous, le <em>Hub</em> a un modèle qui fait exactement cela ! Chargeons-le en utilisant la fonction <code>pipeline()</code> :</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> pipeline

classifier = pipeline(
    <span class="hljs-string">&quot;audio-classification&quot;</span>,
    model=<span class="hljs-string">&quot;anton-l/xtreme_s_xlsr_300m_minds14&quot;</span>,
)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-b3mk4t">Ce pipeline attend les données audio sous forme de tableau NumPy. Tout le prétraitement des données audio brutes sera commodément géré pour nous par le pipeline. Choisissons un exemple pour l’essayer :</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->example = minds[<span class="hljs-number">0</span>]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1tkv2">Si vous vous souvenez de la structure du jeu de données, les données audio brutes sont stockées dans un tableau NumPy sous <code>[&quot;audio&quot;][&quot;array&quot;]</code>, passons-les directement au <code>classifier</code> :</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->classifier(example[<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>])<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-dbzevp"><strong>Sortie :</strong></p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->[
    {<span class="hljs-string">&quot;score&quot;</span>: <span class="hljs-number">0.9631525278091431</span>, <span class="hljs-string">&quot;label&quot;</span>: <span class="hljs-string">&quot;pay_bill&quot;</span>},
    {<span class="hljs-string">&quot;score&quot;</span>: <span class="hljs-number">0.02819698303937912</span>, <span class="hljs-string">&quot;label&quot;</span>: <span class="hljs-string">&quot;freeze&quot;</span>},
    {<span class="hljs-string">&quot;score&quot;</span>: <span class="hljs-number">0.0032787492964416742</span>, <span class="hljs-string">&quot;label&quot;</span>: <span class="hljs-string">&quot;card_issues&quot;</span>},
    {<span class="hljs-string">&quot;score&quot;</span>: <span class="hljs-number">0.0019414445850998163</span>, <span class="hljs-string">&quot;label&quot;</span>: <span class="hljs-string">&quot;abroad&quot;</span>},
    {<span class="hljs-string">&quot;score&quot;</span>: <span class="hljs-number">0.0008378693601116538</span>, <span class="hljs-string">&quot;label&quot;</span>: <span class="hljs-string">&quot;high_value_payment&quot;</span>},
]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-w8dulx">Le modèle est très confiant que l’appelant avait l’intention d’apprendre à payer sa facture. Voyons quelle est l’étiquette réelle pour cet exemple:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->id2label = minds.features[<span class="hljs-string">&quot;intent_class&quot;</span>].int2str
id2label(example[<span class="hljs-string">&quot;intent_class&quot;</span>])<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-dbzevp"><strong>Sortie :</strong></p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-string">&quot;pay_bill&quot;</span><!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-134o58x">L’étiquette prédite est correcte ! Ici, nous avons eu la chance de trouver un modèle capable de classer les étiquettes exactes dont nous avons besoin.
Souvent, lorsqu’il s’agit d’une tâche de classification, l’ensemble de classes d’un modèle pré-entraîné n’est pas exactement le même que les classes que vous devez distinguer par le modèle.
Dans ce cas, vous pouvez finetuner un modèle pré-entraîné pour le « calibrer » en fonction de votre ensemble exact d’étiquettes de classe. Nous apprendrons comment le faire dans les prochaines unités.</p> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/audio-transformers-course/blob/main/chapters/fr/chapter2/audio_classification_pipeline.mdx" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg> <span data-svelte-h="svelte-zjs2n5"><span class="underline">Update</span> on GitHub</span></a>  <p></p> 
			
			<script>
				{
					__sveltekit_oc6cu6 = {
						assets: "/docs/audio-course/pr_239/fr",
						base: "/docs/audio-course/pr_239/fr",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/audio-course/pr_239/fr/_app/immutable/entry/start.48d22ca3.js"),
						import("/docs/audio-course/pr_239/fr/_app/immutable/entry/app.4e1ef9b0.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 13],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>