Skip to content

M
musicinformationretrieval-com
Commit 43a3a5d2 authored by Steve Tjoa's avatar Steve Tjoa
Browse files
Options

onset segmentation

parent 39dd5d79
Show whitespace changes
Inline Side-by-side
Showing with 54 additions and 48 deletions
onset_segmentation.html
View file @ 43a3a5d2
......@@ -11802,6 +11802,16 @@ div#notebook {
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h1 id="Onset-based-Segmentation-with-Backtracking">Onset-based Segmentation with Backtracking<a class="anchor-link" href="#Onset-based-Segmentation-with-Backtracking">&#194;&#182;</a></h1>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p><a href="http://librosa.github.io/librosa/generated/librosa.onset.onset_detect.html#librosa.onset.onset_detect"><code>librosa.onset.onset_detect</code></a> works by finding peaks in a spectral novelty function. However, these peaks may not actually coincide with the initial rise in energy or how we perceive the beginning of a musical note.</p>
<p>The optional keyword parameter <code>backtrack=True</code> will backtrack from each peak to a preceding local minimum. Backtracking can be useful for finding segmentation points such that the onset occurs shortly after the beginning of the segment. We will use <code>backtrack=True</code> to perform onset-based segmentation of a signal.</p>
</div>
</div>
</div>
......@@ -11896,16 +11906,6 @@ div#notebook {
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="librosa.onset.onset_detect"><code>librosa.onset.onset_detect</code><a class="anchor-link" href="#librosa.onset.onset_detect">&#194;&#182;</a></h2>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p><a href="http://librosa.github.io/librosa/generated/librosa.onset.onset_detect.html#librosa.onset.onset_detect"><code>librosa.onset.onset_detect</code></a> works by finding peaks in a spectral novelty function. However, these peaks may not actually coincide with the initial rise in energy or how we perceive the beginning of a musical note.</p>
<p>The optional keyword parameter <code>backtrack=True</code> will backtrack from each peak to a preceding local minimum. Backtracking can be useful for finding segmentation points such that the onset occurs shortly after the beginning of the segment. We will use <code>backtrack=True</code> to perform onset-based segmentation of a signal.</p>
<p>Compute the frame indices for estimated onsets in a signal:</p>
</div>
......@@ -12069,7 +12069,7 @@ div#notebook {
<div class="output_text output_subarea output_execute_result">
<pre>&lt;matplotlib.collections.LineCollection at 0x10ff5dc10&gt;</pre>
<pre>&lt;matplotlib.collections.LineCollection at 0x1150e5490&gt;</pre>
</div>
</div>
......@@ -15270,7 +15270,7 @@ AElFTkSuQmCC
<div class="text_cell_render border-box-sizing rendered_html">
<p>Let's listen to these segments. We will create a function to do the following:</p>
<ol>
<li>Divide the signal into segments beginning at each onset.</li>
<li>Divide the signal into segments beginning at each detected onset.</li>
<li>Pad each segment with 500 ms of silence.</li>
<li>Concatenate the padded segments.</li>
</ol>
......@@ -15285,10 +15285,10 @@ AElFTkSuQmCC
<div class="input_area">
<div class=" highlight hl-ipython2"><pre><span></span><span class="k">def</span> <span class="nf">concatenate_segments</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">onset_samples</span><span class="p">,</span> <span class="n">pad_duration</span><span class="o">=</span><span class="mf">0.500</span><span class="p">):</span>
<span class="sd">&quot;&quot;&quot;Concatenate segments into one signal.&quot;&quot;&quot;</span>
<span class="n">silence</span> <span class="o">=</span> <span class="n">numpy</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">pad_duration</span><span class="o">*</span><span class="n">sr</span><span class="p">))</span>
<span class="n">frame_sz</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">numpy</span><span class="o">.</span><span class="n">diff</span><span class="p">(</span><span class="n">onset_samples</span><span class="p">))</span>
<span class="n">silence</span> <span class="o">=</span> <span class="n">numpy</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">pad_duration</span><span class="o">*</span><span class="n">sr</span><span class="p">))</span> <span class="c1"># silence</span>
<span class="n">frame_sz</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">numpy</span><span class="o">.</span><span class="n">diff</span><span class="p">(</span><span class="n">onset_samples</span><span class="p">))</span> <span class="c1"># every segment has uniform frame size</span>
<span class="k">return</span> <span class="n">numpy</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span>
<span class="n">numpy</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">x</span><span class="p">[</span><span class="n">i</span><span class="p">:</span><span class="n">i</span><span class="o">+</span><span class="n">frame_sz</span><span class="p">],</span> <span class="n">silence</span><span class="p">])</span>
<span class="n">numpy</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">x</span><span class="p">[</span><span class="n">i</span><span class="p">:</span><span class="n">i</span><span class="o">+</span><span class="n">frame_sz</span><span class="p">],</span> <span class="n">silence</span><span class="p">])</span> <span class="c1"># pad segment with silence</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">onset_samples</span>
<span class="p">])</span>
</pre></div>
......@@ -15379,7 +15379,7 @@ AElFTkSuQmCC
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Backtracking">Backtracking<a class="anchor-link" href="#Backtracking">&#194;&#182;</a></h2>
<h2 id="librosa.onset.onset_backtrack"><code>librosa.onset.onset_backtrack</code><a class="anchor-link" href="#librosa.onset.onset_backtrack">&#194;&#182;</a></h2>
</div>
</div>
</div>
......@@ -15387,7 +15387,9 @@ AElFTkSuQmCC
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>We can avoid this glitch by backtracking:</p>
<p>We can avoid this glitch by backtracking from the detected onsets.</p>
<p>When setting the parameter <code>backtrack=True</code>, <code>librosa.onset.onset_detect</code> will call <a href="http://librosa.github.io/librosa/generated/librosa.onset.onset_backtrack.html"><code>librosa.onset.onset_backtrack</code></a>.
For each detected onset, <code>librosa.onset.onset_backtrack</code> searches backward for a local minimum.</p>
</div>
</div>
......@@ -15485,7 +15487,7 @@ AElFTkSuQmCC
<div class="output_text output_subarea output_execute_result">
<pre>&lt;matplotlib.collections.LineCollection at 0x1103e2810&gt;</pre>
<pre>&lt;matplotlib.collections.LineCollection at 0x114cef190&gt;</pre>
</div>
</div>
......@@ -18731,7 +18733,7 @@ IiLKCG8eiIiIiIgoI7x5ICIiIiKijPDmgYiIiIiIMvI/mbmRboJD3V0AAAAASUVORK5CYII=
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>As we can hear, now the segments are perfectly segmented.</p>
<p>While listening, notice now the segments are perfectly segmented.</p>
</div>
</div>
onset_segmentation.ipynb
View file @ 43a3a5d2
This source diff could not be displayed because it is too large. You can view the blob instead.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment