Hugging Face's logo

Buckets:
hf-doc-build
/
doc-dev

Files
xet
hf-doc-build/doc-dev / transformers /pr_33892 /en /feature_extractors.html
rtrm's picture
rtrm
download
raw
41.6 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Feature extractors&quot;,&quot;local&quot;:&quot;feature-extractors&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Feature extractor classes&quot;,&quot;local&quot;:&quot;feature-extractor-classes&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Preprocess&quot;,&quot;local&quot;:&quot;preprocess&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Padding&quot;,&quot;local&quot;:&quot;padding&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Truncation&quot;,&quot;local&quot;:&quot;truncation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Resampling&quot;,&quot;local&quot;:&quot;resampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2}],&quot;depth&quot;:1}">
 <link href="/docs/transformers/pr_33892/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/entry/start.b2c4257a.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/scheduler.31fdf58d.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/singletons.9860629f.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/index.252883d5.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/paths.e85c0ec8.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/entry/app.05ef1f97.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/preload-helper.40847a0e.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/index.2f76fdf0.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/nodes/0.ca4aafa4.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/each.e59479a4.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/nodes/24.f2bbbf0a.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/CopyLLMTxtMenu.ff482081.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.71f274cc.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/IconCopy.ac192424.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/CodeBlock.ab12f8e1.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33892/en/_app/immutable/chunks/HfOption.fb051768.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Feature extractors&quot;,&quot;local&quot;:&quot;feature-extractors&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Feature extractor classes&quot;,&quot;local&quot;:&quot;feature-extractor-classes&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Preprocess&quot;,&quot;local&quot;:&quot;preprocess&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Padding&quot;,&quot;local&quot;:&quot;padding&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Truncation&quot;,&quot;local&quot;:&quot;truncation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Resampling&quot;,&quot;local&quot;:&quot;resampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2}],&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 max-sm:gap-0.5 h-6 max-sm:h-5 px-2 max-sm:px-1.5 text-[11px] max-sm:text-[9px] 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"><svg class="w-3 h-3 max-sm:w-2.5 max-sm:h-2.5" 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-6 max-sm:h-5 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 w-3 h-3 max-sm:w-2.5 max-sm:h-2.5 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="feature-extractors" 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="#feature-extractors"><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>Feature extractors</span></h1> <p data-svelte-h="svelte-11npljq">Feature extractors preprocess audio data into the correct format for a given model. It takes the raw audio signal and converts it into a tensor that can be fed to a model. The tensor shape depends on the model, but the feature extractor will correctly preprocess the audio data for you given the model you’re using. Feature extractors also include methods for padding, truncation, and resampling.</p> <p data-svelte-h="svelte-1jf1nkn">Call <a href="/docs/transformers/pr_33892/en/model_doc/auto#transformers.AutoFeatureExtractor.from_pretrained">from_pretrained()</a> to load a feature extractor and its preprocessor configuration from the Hugging Face <a href="https://hf.co/models" rel="nofollow">Hub</a> or local directory. The feature extractor and preprocessor configuration is saved in a <a href="https://hf.co/openai/whisper-tiny/blob/main/preprocessor_config.json" rel="nofollow">preprocessor_config.json</a> file.</p> <p data-svelte-h="svelte-t8kulv">Pass the audio signal, typically stored in <code>array</code>, to the feature extractor and set the <code>sampling_rate</code> parameter to the pretrained audio models sampling rate. It is important the sampling rate of the audio data matches the sampling rate of the data a pretrained audio model was trained on.</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> AutoFeatureExtractor
feature_extractor = AutoFeatureExtractor.from_pretrained(<span class="hljs-string">&quot;facebook/wav2vec2-base&quot;</span>)
dataset = load_dataset(<span class="hljs-string">&quot;PolyAI/minds14&quot;</span>, name=<span class="hljs-string">&quot;en-US&quot;</span>, split=<span class="hljs-string">&quot;train&quot;</span>)
processed_sample = feature_extractor(dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>], sampling_rate=<span class="hljs-number">16000</span>)
processed_sample
{<span class="hljs-string">&#x27;input_values&#x27;</span>: [array([ <span class="hljs-number">9.4472744e-05</span>, <span class="hljs-number">3.0777880e-03</span>, -<span class="hljs-number">2.8888427e-03</span>, ...,
 -<span class="hljs-number">2.8888427e-03</span>, <span class="hljs-number">9.4472744e-05</span>, <span class="hljs-number">9.4472744e-05</span>], dtype=float32)]}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1cuyph">The feature extractor returns an input, <code>input_values</code>, that is ready for the model to consume.</p> <p data-svelte-h="svelte-xmmv9">This guide walks you through the feature extractor classes and how to preprocess audio data.</p> <h2 class="relative group"><a id="feature-extractor-classes" 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="#feature-extractor-classes"><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>Feature extractor classes</span></h2> <p data-svelte-h="svelte-46iwh5">Transformers feature extractors inherit from the base <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.SequenceFeatureExtractor">SequenceFeatureExtractor</a> class which subclasses <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.FeatureExtractionMixin">FeatureExtractionMixin</a>.</p> <ul data-svelte-h="svelte-1nj58d0"><li><a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.SequenceFeatureExtractor">SequenceFeatureExtractor</a> provides a method to <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.SequenceFeatureExtractor.pad">pad()</a> sequences to a certain length to avoid uneven sequence lengths.</li> <li><a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.FeatureExtractionMixin">FeatureExtractionMixin</a> provides <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.FeatureExtractionMixin.from_pretrained">from_pretrained()</a> and <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.FeatureExtractionMixin.save_pretrained">save_pretrained()</a> to load and save a feature extractor.</li></ul> <p data-svelte-h="svelte-19ppiwj">There are two ways you can load a feature extractor, <a href="/docs/transformers/pr_33892/en/model_doc/auto#transformers.AutoFeatureExtractor">AutoFeatureExtractor</a> and a model-specific feature extractor class.</p> <div class="flex space-x-2 items-center my-1.5 mr-8 h-7 !pl-0 -mx-3 md:mx-0"><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd border-gray-800 bg-black dark:bg-gray-700 text-white">AutoFeatureExtractor </div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">model-specific feature extractor </div></div> <div class="language-select"><p data-svelte-h="svelte-1gxot7v">The <a href="./model_doc/auto">AutoClass</a> API automatically loads the correct feature extractor for a given model.</p> <p data-svelte-h="svelte-ses8hi">Use <a href="/docs/transformers/pr_33892/en/model_doc/auto#transformers.AutoFeatureExtractor.from_pretrained">from_pretrained()</a> to load a feature extractor.</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> AutoFeatureExtractor
feature_extractor = AutoFeatureExtractor.from_pretrained(<span class="hljs-string">&quot;openai/whisper-tiny&quot;</span>)<!-- HTML_TAG_END --></pre></div> </div> <h2 class="relative group"><a id="preprocess" 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="#preprocess"><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>Preprocess</span></h2> <p data-svelte-h="svelte-f4il9r">A feature extractor expects the input as a PyTorch tensor of a certain shape. The exact input shape can vary depending on the specific audio model you’re using.</p> <p data-svelte-h="svelte-1xo1c8p">For example, <a href="https://huggingface.co/docs/transformers/model_doc/whisper" rel="nofollow">Whisper</a> expects <code>input_features</code> to be a tensor of shape <code>(batch_size, feature_size, sequence_length)</code> but <a href="https://hf.co/docs/transformers/model_doc/wav2vec2" rel="nofollow">Wav2Vec2</a> expects <code>input_values</code> to be a tensor of shape <code>(batch_size, sequence_length)</code>.</p> <p data-svelte-h="svelte-sizsez">The feature extractor generates the correct input shape for whichever audio model you’re using.</p> <p data-svelte-h="svelte-81ch6r">A feature extractor also sets the sampling rate (the number of audio signal values taken per second) of the audio files. The sampling rate of your audio data must match the sampling rate of the dataset a pretrained model was trained on. This value is typically given in the model card.</p> <p data-svelte-h="svelte-k6cln2">Load a dataset and feature extractor with <a href="/docs/transformers/pr_33892/en/main_classes/feature_extractor#transformers.FeatureExtractionMixin.from_pretrained">from_pretrained()</a>.</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, Audio
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoFeatureExtractor
dataset = load_dataset(<span class="hljs-string">&quot;PolyAI/minds14&quot;</span>, name=<span class="hljs-string">&quot;en-US&quot;</span>, split=<span class="hljs-string">&quot;train&quot;</span>)
feature_extractor = AutoFeatureExtractor.from_pretrained(<span class="hljs-string">&quot;facebook/wav2vec2-base&quot;</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-10j4ocw">Check out the first example from the dataset and access the <code>audio</code> column which contains <code>array</code>, the raw audio signal.</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 -->dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>]
array([ <span class="hljs-number">0.</span> , <span class="hljs-number">0.00024414</span>, -<span class="hljs-number">0.00024414</span>, ..., -<span class="hljs-number">0.00024414</span>,
 <span class="hljs-number">0.</span> , <span class="hljs-number">0.</span> ])<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1tguczj">The feature extractor preprocesses <code>array</code> into the expected input format for a given audio model. Use the <code>sampling_rate</code> parameter to set the appropriate sampling rate.</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 -->processed_dataset = feature_extractor(dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>], sampling_rate=<span class="hljs-number">16000</span>)
processed_dataset
{<span class="hljs-string">&#x27;input_values&#x27;</span>: [array([ <span class="hljs-number">9.4472744e-05</span>, <span class="hljs-number">3.0777880e-03</span>, -<span class="hljs-number">2.8888427e-03</span>, ...,
 -<span class="hljs-number">2.8888427e-03</span>, <span class="hljs-number">9.4472744e-05</span>, <span class="hljs-number">9.4472744e-05</span>], dtype=float32)]}<!-- HTML_TAG_END --></pre></div> <h3 class="relative group"><a id="padding" 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="#padding"><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>Padding</span></h3> <p data-svelte-h="svelte-1m4vi3k">Audio sequence lengths that are different is an issue because Transformers expects all sequences to have the same lengths so they can be batched. Uneven sequence lengths can’t be batched.</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 -->dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>].shape
(<span class="hljs-number">86699</span>,)
dataset[<span class="hljs-number">1</span>][<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-string">&quot;array&quot;</span>].shape
(<span class="hljs-number">53248</span>,)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1iybb0r">Padding adds a special <em>padding token</em> to ensure all sequences have the same length. The feature extractor adds a <code>0</code> - interpreted as silence - to <code>array</code> to pad it. Set <code>padding=True</code> to pad sequences to the longest sequence length in the batch.</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">def</span> <span class="hljs-title function_">preprocess_function</span>(<span class="hljs-params">examples</span>):
 audio_arrays = [x[<span class="hljs-string">&quot;array&quot;</span>] <span class="hljs-keyword">for</span> x <span class="hljs-keyword">in</span> examples[<span class="hljs-string">&quot;audio&quot;</span>]]
 inputs = feature_extractor(
 audio_arrays,
 sampling_rate=<span class="hljs-number">16000</span>,
 padding=<span class="hljs-literal">True</span>,
 )
 <span class="hljs-keyword">return</span> inputs
processed_dataset = preprocess_function(dataset[:<span class="hljs-number">5</span>])
processed_dataset[<span class="hljs-string">&quot;input_values&quot;</span>][<span class="hljs-number">0</span>].shape
(<span class="hljs-number">86699</span>,)
processed_dataset[<span class="hljs-string">&quot;input_values&quot;</span>][<span class="hljs-number">1</span>].shape
(<span class="hljs-number">86699</span>,)<!-- HTML_TAG_END --></pre></div> <h3 class="relative group"><a id="truncation" 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="#truncation"><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>Truncation</span></h3> <p data-svelte-h="svelte-fkz01s">Models can only process sequences up to a certain length before crashing.</p> <p data-svelte-h="svelte-sxi7vo">Truncation is a strategy for removing excess tokens from a sequence to ensure it doesn’t exceed the maximum length. Set <code>truncation=True</code> to truncate a sequence to the length in the <code>max_length</code> parameter.</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">def</span> <span class="hljs-title function_">preprocess_function</span>(<span class="hljs-params">examples</span>):
 audio_arrays = [x[<span class="hljs-string">&quot;array&quot;</span>] <span class="hljs-keyword">for</span> x <span class="hljs-keyword">in</span> examples[<span class="hljs-string">&quot;audio&quot;</span>]]
 inputs = feature_extractor(
 audio_arrays,
 sampling_rate=<span class="hljs-number">16000</span>,
 max_length=<span class="hljs-number">50000</span>,
 truncation=<span class="hljs-literal">True</span>,
 )
 <span class="hljs-keyword">return</span> inputs
processed_dataset = preprocess_function(dataset[:<span class="hljs-number">5</span>])
processed_dataset[<span class="hljs-string">&quot;input_values&quot;</span>][<span class="hljs-number">0</span>].shape
(<span class="hljs-number">50000</span>,)
processed_dataset[<span class="hljs-string">&quot;input_values&quot;</span>][<span class="hljs-number">1</span>].shape
(<span class="hljs-number">50000</span>,)<!-- HTML_TAG_END --></pre></div> <h3 class="relative group"><a id="resampling" 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="#resampling"><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>Resampling</span></h3> <p data-svelte-h="svelte-1tbpz6p">The <a href="https://hf.co/docs/datasets/index" rel="nofollow">Datasets</a> library can also resample audio data to match an audio models expected sampling rate. This method resamples the audio data on the fly when they’re loaded which can be faster than resampling the entire dataset in-place.</p> <p data-svelte-h="svelte-8axfxt">The audio dataset you’ve been working on has a sampling rate of 8kHz and the pretrained model expects 16kHz.</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 -->dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>]
{<span class="hljs-string">&#x27;path&#x27;</span>: <span class="hljs-string">&#x27;/root/.cache/huggingface/datasets/downloads/extracted/f507fdca7f475d961f5bb7093bcc9d544f16f8cab8608e772a2ed4fbeb4d6f50/en-US~JOINT_ACCOUNT/602ba55abb1e6d0fbce92065.wav&#x27;</span>,
 <span class="hljs-string">&#x27;array&#x27;</span>: array([ <span class="hljs-number">0.</span> , <span class="hljs-number">0.00024414</span>, -<span class="hljs-number">0.00024414</span>, ..., -<span class="hljs-number">0.00024414</span>,
 <span class="hljs-number">0.</span> , <span class="hljs-number">0.</span> ]),
 <span class="hljs-string">&#x27;sampling_rate&#x27;</span>: <span class="hljs-number">8000</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-48tw20">Call <a href="https://huggingface.co/docs/datasets/main/en/package_reference/main_classes#datasets.Dataset.cast_column" rel="nofollow">cast_column</a> on the <code>audio</code> column to upsample the sampling rate to 16kHz.</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 -->dataset = dataset.cast_column(<span class="hljs-string">&quot;audio&quot;</span>, Audio(sampling_rate=<span class="hljs-number">16000</span>))<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-57e76x">When you load the dataset sample, it is now resampled to 16kHz.</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 -->dataset[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;audio&quot;</span>]
{<span class="hljs-string">&#x27;path&#x27;</span>: <span class="hljs-string">&#x27;/root/.cache/huggingface/datasets/downloads/extracted/f507fdca7f475d961f5bb7093bcc9d544f16f8cab8608e772a2ed4fbeb4d6f50/en-US~JOINT_ACCOUNT/602ba55abb1e6d0fbce92065.wav&#x27;</span>,
 <span class="hljs-string">&#x27;array&#x27;</span>: array([ <span class="hljs-number">1.70562416e-05</span>, <span class="hljs-number">2.18727451e-04</span>, <span class="hljs-number">2.28099874e-04</span>, ...,
 <span class="hljs-number">3.43842403e-05</span>, -<span class="hljs-number">5.96364771e-06</span>, -<span class="hljs-number">1.76846661e-05</span>]),
 <span class="hljs-string">&#x27;sampling_rate&#x27;</span>: <span class="hljs-number">16000</span>}<!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/en/feature_extractors.md" 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_16tnnm8 = {
 assets: "/docs/transformers/pr_33892/en",
 base: "/docs/transformers/pr_33892/en",
 env: {}
 };
 const element = document.currentScript.parentElement;
 const data = [null,null];
 Promise.all([
 import("/docs/transformers/pr_33892/en/_app/immutable/entry/start.b2c4257a.js"),
 import("/docs/transformers/pr_33892/en/_app/immutable/entry/app.05ef1f97.js")
 ]).then(([kit, app]) => {
 kit.start(app, element, {
 node_ids: [0, 24],
 data,
 form: null,
 error: null
 });
 });
 }
 </script>

Xet Storage Details

Size:
41.6 kB
·
Xet hash:
d256a1bc61fdce1401cbd3dad9b2806ab59bb8dd33fe14359dd90d4caed7c653

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.