|
<!DOCTYPE html> |
|
<html> |
|
<head> |
|
<meta charset="utf-8" /> |
|
<meta name="viewport" content="width=device-width" /> |
|
<title>ARCH: Audio Representation benCHmark</title> |
|
<link href='http://fonts.googleapis.com/css?family=Roboto' rel='stylesheet' type='text/css'> |
|
<link rel="stylesheet" href="style.css" /> |
|
</head> |
|
<body> |
|
|
|
<img src="arch_logo.png" class="center_img width500"> |
|
|
|
<br> |
|
|
|
<p style="text-align: center;"> |
|
ARCH is a framework designed to benchmark audio representations. The goal is to provide a unified framework for researchers to compare their audio representations and to provide a benchmark for the community to evaluate their models. |
|
The project is currently in its first release. The details about the datasets and the models are available in the <a href="https://github.com/MorenoLaQuatra/ARCH/" target="_blank">GitHub repository</a>. |
|
</p> |
|
|
|
<br><br> |
|
|
|
<h2 style="text-align: center;">Results on the ARCH benchmark - Version 1.0</h2> |
|
<style type="text/css"> |
|
.bottom_margin { |
|
|
|
border-bottom: 1px solid #000000; |
|
} |
|
.tg {border-collapse:collapse;border-color:#ccc;border-spacing:0;border-style:solid;border-width:1px;} |
|
.tg td{background-color:#fff;border-color:#ccc;border-style:solid;border-width:0px;color:#333; |
|
font-family:Arial, sans-serif;font-size:14px;overflow:hidden;padding:10px 5px;word-break:normal;} |
|
.tg th{background-color:#f0f0f0;border-color:#ccc;border-style:solid;border-width:0px;color:#333; |
|
font-family:Arial, sans-serif;font-size:14px;font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} |
|
.tg .tg-r0pt{background-color:#cbcefb;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-2ujd{background-color:#cbcefb;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-lqy8{background-color:#cbcefb;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-bm9l{background-color:#cbcefb;border-color:#000000;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-ei74{background-color:#ffce93;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-tcrt{font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-x5s5{font-family:"Lucida Console", Monaco, monospace !important;text-align:center;vertical-align:top} |
|
.tg .tg-98fj{background-color:#ffce93;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-dpyv{background-color:#ce6301;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-3zje{background-color:#ce6301;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-hezg{background-color:#ffce93;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-7yg0{background-color:#ffccc9;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-svvj{background-color:#f9f9f9;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-ttzb{background-color:#ffce93;border-color:#000000;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-81gd{background-color:#cbcefb;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-ibhv{background-color:#ffccc9;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-hv44{border-color:inherit;font-family:"Lucida Console", Monaco, monospace !important;text-align:center;vertical-align:top} |
|
.tg .tg-w16u{background-color:#cbcefb;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-imml{background-color:#ffce93;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-th5u{background-color:#c0c0c0;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-j4yz{background-color:#fd6864;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-d1sc{background-color:#303498;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-qgag{background-color:#fd6864;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-ge3d{background-color:#303498;border-color:inherit;color:#ffffff;font-family:Arial, Helvetica, sans-serif !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-lcf4{border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-22k7{background-color:#ffccc9;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-jqvv{background-color:#f9f9f9;border-color:inherit;font-family:"Lucida Console", Monaco, monospace !important; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-13no{background-color:#ffccc9;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-heot{background-color:#cbcefb;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-6tzy{background-color:#ffce93;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-1lb8{border-color:#000000;font-family:"Lucida Console", Monaco, monospace !important;text-align:center;vertical-align:top} |
|
.tg .tg-0qvc{border-color:#000000;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-c78u{background-color:#ffccc9;border-color:#000000;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
.tg .tg-i8q9{background-color:#ffccc9;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-hh0f{background-color:#ffccc9;border-color:#000000;font-family:Arial, Helvetica, sans-serif !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-ncmw{background-color:#cbcefb;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-n3t2{background-color:#ffce93;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-zugs{background-color:#f9f9f9;font-family:"Lucida Console", Monaco, monospace !important;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-fb7a{background-color:#f9f9f9;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-n84z{background-color:#ffccc9;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-mekq{background-color:#ffccc9;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold;text-align:center; |
|
vertical-align:top} |
|
.tg .tg-t89l{background-color:#ffce93;font-family:Arial, Helvetica, sans-serif !important;text-align:center;vertical-align:top} |
|
.tg .tg-wssm{background-color:#ffccc9;border-color:inherit;font-family:Arial, Helvetica, sans-serif !important;font-weight:bold; |
|
text-align:center;vertical-align:top} |
|
</style> |
|
<table class="tg center_table"> |
|
<thead> |
|
<tr> |
|
<th class="tg-th5u" rowspan="2">Model</th> |
|
<th class="tg-th5u" rowspan="2"> Size </th> |
|
<th class="tg-j4yz" colspan="4">Sound</th> |
|
<th class="tg-d1sc" colspan="4">Music</th> |
|
<th class="tg-dpyv" colspan="4">Speech</th> |
|
</tr> |
|
<tr> |
|
<th class="tg-qgag"> ESC-50 </th> |
|
<th class="tg-qgag"> US8K </th> |
|
<th class="tg-qgag"> FSD50K </th> |
|
<th class="tg-qgag"> VIVAE </th> |
|
<th class="tg-ge3d"> FMA </th> |
|
<th class="tg-ge3d"> MTT </th> |
|
<th class="tg-ge3d"> IRMAS </th> |
|
<th class="tg-ge3d"> MS-DB </th> |
|
<th class="tg-3zje"> RAVDESS </th> |
|
<th class="tg-3zje"> A-MNIST </th> |
|
<th class="tg-3zje"> SLURP </th> |
|
<th class="tg-3zje"> EMOVO </th> |
|
</tr> |
|
</thead> |
|
<tbody> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/facebook/wav2vec2-base" target="_blank" rel="noopener noreferrer">facebook/wav2vec2-base</a></td> |
|
<td class="tg-lcf4">B</td> |
|
<td class="tg-22k7">45.73</td> |
|
<td class="tg-22k7">55.48</td> |
|
<td class="tg-22k7">19.39</td> |
|
<td class="tg-22k7">31.47</td> |
|
<td class="tg-lqy8">50.54</td> |
|
<td class="tg-lqy8">37.56</td> |
|
<td class="tg-lqy8">35.14</td> |
|
<td class="tg-lqy8">66.06</td> |
|
<td class="tg-98fj">55.32</td> |
|
<td class="tg-98fj">86.38</td> |
|
<td class="tg-98fj">14.37</td> |
|
<td class="tg-98fj">31.80</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/microsoft/wavlm-base" target="_blank" rel="noopener noreferrer">microsoft/wavlm-base</a></td> |
|
<td class="tg-svvj">B</td> |
|
<td class="tg-13no">49.88</td> |
|
<td class="tg-13no">61.84</td> |
|
<td class="tg-13no">17.63</td> |
|
<td class="tg-13no">36.31</td> |
|
<td class="tg-heot">48.71</td> |
|
<td class="tg-heot">34.93</td> |
|
<td class="tg-heot">32.62</td> |
|
<td class="tg-heot">54.18</td> |
|
<td class="tg-imml"><span style="font-style:normal">67.94</span></td> |
|
<td class="tg-6tzy">99.50</td> |
|
<td class="tg-6tzy">30.98</td> |
|
<td class="tg-imml">43.08</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/microsoft/wavlm-base-plus" target="_blank" rel="noopener noreferrer">microsoft/wavlm-base-plus</a></td> |
|
<td class="tg-lcf4">B</td> |
|
<td class="tg-22k7">58.73</td> |
|
<td class="tg-22k7">64.07</td> |
|
<td class="tg-22k7">21.57</td> |
|
<td class="tg-22k7">36.17</td> |
|
<td class="tg-lqy8">56.17</td> |
|
<td class="tg-lqy8">38.24</td> |
|
<td class="tg-lqy8">35.76</td> |
|
<td class="tg-lqy8">57.51</td> |
|
<td class="tg-98fj">52.20</td> |
|
<td class="tg-ei74">99.63</td> |
|
<td class="tg-98fj">28.06</td> |
|
<td class="tg-98fj">36.73</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/facebook/hubert-base-ls960" target="_blank" rel="noopener noreferrer">facebook/hubert-base-ls960</a></td> |
|
<td class="tg-svvj">B</td> |
|
<td class="tg-13no">58.90</td> |
|
<td class="tg-13no">67.28</td> |
|
<td class="tg-13no">24.53</td> |
|
<td class="tg-ibhv">40.48</td> |
|
<td class="tg-heot">54.63</td> |
|
<td class="tg-heot">38.78</td> |
|
<td class="tg-heot">36.65</td> |
|
<td class="tg-heot">58.46</td> |
|
<td class="tg-6tzy">65.28</td> |
|
<td class="tg-6tzy">99.58</td> |
|
<td class="tg-6tzy">33.75</td> |
|
<td class="tg-6tzy">40.48</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/facebook/data2vec-audio-base" target="_blank" rel="noopener noreferrer">facebook/data2vec-audio-base</a></td> |
|
<td class="tg-lcf4">B</td> |
|
<td class="tg-22k7">23.63</td> |
|
<td class="tg-22k7">45.63</td> |
|
<td class="tg-22k7">10.06</td> |
|
<td class="tg-22k7">30.19</td> |
|
<td class="tg-lqy8">40.58</td> |
|
<td class="tg-lqy8">27.60</td> |
|
<td class="tg-lqy8">25.87</td> |
|
<td class="tg-lqy8">50.74</td> |
|
<td class="tg-98fj">48.03</td> |
|
<td class="tg-98fj">99.06</td> |
|
<td class="tg-ei74">43.57</td> |
|
<td class="tg-98fj">27.27</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/ALM/wav2vec2-base-audioset" target="_blank" rel="noopener noreferrer">ALM/wav2vec2-base-audioset</a></td> |
|
<td class="tg-svvj">B</td> |
|
<td class="tg-13no">52.61</td> |
|
<td class="tg-13no">70.48</td> |
|
<td class="tg-13no"><span style="font-weight:400;font-style:normal">21.29</span></td> |
|
<td class="tg-13no">31.26</td> |
|
<td class="tg-heot">59.50</td> |
|
<td class="tg-heot"><span style="font-weight:400;font-style:normal">37.92</span></td> |
|
<td class="tg-heot">35.85</td> |
|
<td class="tg-heot">64.61</td> |
|
<td class="tg-6tzy">45.94</td> |
|
<td class="tg-6tzy">88.09 </td> |
|
<td class="tg-6tzy">11.00</td> |
|
<td class="tg-6tzy">3<span style="font-weight:400;font-style:normal">0.83</span></td> |
|
</tr> |
|
<tr class="bottom_margin"> |
|
<td class="tg-1lb8"><a href="https://huggingface.co/ALM/hubert-base-audioset" target="_blank" rel="noopener noreferrer">ALM/hubert-base-audioset</a></td> |
|
<td class="tg-0qvc">B</td> |
|
<td class="tg-c78u">68.80</td> |
|
<td class="tg-i8q9"><span style="font-weight:700;font-style:normal"><u></span><span style="font-style:normal">79.09</span><span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-c78u"><span style="font-style:normal">31.05</span></td> |
|
<td class="tg-hh0f">40.06</td> |
|
<td class="tg-bm9l"><span style="font-style:normal">65.87</span></td> |
|
<td class="tg-bm9l"><span style="font-style:normal">43.44</span></td> |
|
<td class="tg-bm9l"><span style="font-style:normal">47.67</span></td> |
|
<td class="tg-bm9l">67.81</td> |
|
<td class="tg-ttzb">63.54</td> |
|
<td class="tg-ttzb"><span style="font-weight:400;font-style:normal">98.84</span></td> |
|
<td class="tg-ttzb"><span style="font-weight:400;font-style:normal">20.53</span></td> |
|
<td class="tg-ttzb">33.39</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/facebook/wav2vec2-large-robust" target="_blank" rel="noopener noreferrer">facebook/wav2vec2-large-robust</a></td> |
|
<td class="tg-svvj">L</td> |
|
<td class="tg-13no">13.13</td> |
|
<td class="tg-13no">42.70</td> |
|
<td class="tg-13no">5.80</td> |
|
<td class="tg-13no">22.01</td> |
|
<td class="tg-heot">41.71</td> |
|
<td class="tg-heot">20.95</td> |
|
<td class="tg-heot">19.91</td> |
|
<td class="tg-heot">50.23</td> |
|
<td class="tg-6tzy">11.57</td> |
|
<td class="tg-6tzy">45.74</td> |
|
<td class="tg-6tzy">7.33</td> |
|
<td class="tg-6tzy">19.27</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/facebook/wav2vec2-xls-r-300m" target="_blank" rel="noopener noreferrer">facebook/wav2vec2-xls-r-300m</a></td> |
|
<td class="tg-lcf4">L</td> |
|
<td class="tg-22k7">51.28</td> |
|
<td class="tg-22k7">69.96</td> |
|
<td class="tg-22k7">23.71</td> |
|
<td class="tg-22k7">36.28</td> |
|
<td class="tg-lqy8">56.96</td> |
|
<td class="tg-lqy8">38.28</td> |
|
<td class="tg-lqy8">38.42</td> |
|
<td class="tg-lqy8">66.71</td> |
|
<td class="tg-98fj">31.48</td> |
|
<td class="tg-98fj">98.88</td> |
|
<td class="tg-98fj">12.74</td> |
|
<td class="tg-98fj">20.35</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/microsoft/wavlm-large" target="_blank" rel="noopener noreferrer">microsoft/wavlm-large</a></td> |
|
<td class="tg-svvj">L</td> |
|
<td class="tg-13no">67.20</td> |
|
<td class="tg-13no">70.92</td> |
|
<td class="tg-13no">32.21</td> |
|
<td class="tg-13no">42.51</td> |
|
<td class="tg-heot">61.13</td> |
|
<td class="tg-heot">41.29</td> |
|
<td class="tg-heot">42.53</td> |
|
<td class="tg-heot">68.00</td> |
|
<td class="tg-6tzy">71.76</td> |
|
<td class="tg-6tzy">99.75</td> |
|
<td class="tg-6tzy">42.34</td> |
|
<td class="tg-imml">45.29</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/facebook/hubert-large-ll60k" target="_blank" rel="noopener noreferrer">facebook/hubert-large-ll60k</a></td> |
|
<td class="tg-lcf4">L</td> |
|
<td class="tg-22k7">63.98</td> |
|
<td class="tg-22k7">70.00</td> |
|
<td class="tg-22k7">29.51</td> |
|
<td class="tg-22k7">40.95</td> |
|
<td class="tg-lqy8">54.79</td> |
|
<td class="tg-lqy8">38.36</td> |
|
<td class="tg-lqy8">36.81</td> |
|
<td class="tg-lqy8">64.08</td> |
|
<td class="tg-98fj">72.57</td> |
|
<td class="tg-ei74">99.95</td> |
|
<td class="tg-ei74">45.26</td> |
|
<td class="tg-98fj">43.76</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/facebook/data2vec-audio-large" target="_blank" rel="noopener noreferrer">facebook/data2vec-audio-large</a></td> |
|
<td class="tg-svvj">L</td> |
|
<td class="tg-13no">25.35</td> |
|
<td class="tg-13no">49.15</td> |
|
<td class="tg-13no">10.82</td> |
|
<td class="tg-13no">30.57</td> |
|
<td class="tg-heot">43.46</td> |
|
<td class="tg-heot">28.52</td> |
|
<td class="tg-heot">27.08</td> |
|
<td class="tg-heot">44.20</td> |
|
<td class="tg-6tzy">45.14</td> |
|
<td class="tg-6tzy">99.15</td> |
|
<td class="tg-6tzy">28.60</td> |
|
<td class="tg-6tzy">23.07</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-x5s5"><a href="https://huggingface.co/ALM/wav2vec2-large-audioset" target="_blank" rel="noopener noreferrer">ALM/wav2vec2-large-audioset</a></td> |
|
<td class="tg-tcrt">L</td> |
|
<td class="tg-i8q9"><span style="font-weight:500;font-style:normal"><u></span>74.39</u></td> |
|
<td class="tg-i8q9"><span style="font-style:normal">79.00</span></td> |
|
<td class="tg-i8q9"><span style="font-weight:700;font-style:normal"><u></span>37.58<span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-7yg0">39.65</td> |
|
<td class="tg-2ujd"><span style="font-weight:400;font-style:normal">66.58</span></td> |
|
<td class="tg-ncmw"><span style="font-weight:700;font-style:normal"><u></span><span style="font-style:normal">44.51</span><span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-2ujd">49.87</td> |
|
<td class="tg-2ujd"><span style="font-style:normal">76.90</span></td> |
|
<td class="tg-n3t2"><span style="font-weight:400;font-style:normal">59.49</span></td> |
|
<td class="tg-n3t2">99.42</td> |
|
<td class="tg-n3t2"><span style="font-weight:400;font-style:normal">17.74</span></td> |
|
<td class="tg-n3t2">38.20</td> |
|
</tr> |
|
<tr class="bottom_margin"> |
|
<td class="tg-zugs"><a href="https://huggingface.co/ALM/hubert-large-audioset" target="_blank" rel="noopener noreferrer">ALM/hubert-large-audioset</a></td> |
|
<td class="tg-fb7a">L</td> |
|
<td class="tg-n84z"><span style="font-weight:400;font-style:normal">71.52</span></td> |
|
<td class="tg-n84z">75.63</td> |
|
<td class="tg-n84z">37.41</td> |
|
<td class="tg-mekq"><span style="font-weight:700;font-style:normal"><u></span>44.28<span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-81gd"><span style="font-weight:700;font-style:normal"><u></span><span style="font-style:normal">67.54</span><span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-w16u"><span style="font-weight:400;font-style:normal">43.35</span></td> |
|
<td class="tg-81gd"><span style="font-weight:700;font-style:normal"><u></span><span style="font-style:normal">50.46</span><span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-81gd"><span style="font-weight:700;font-style:normal"><u></span><span style="font-style:normal">77.82</span><span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-hezg"><span style="font-style:normal">73.26</span></td> |
|
<td class="tg-t89l">99.59</td> |
|
<td class="tg-t89l">20.46</td> |
|
<td class="tg-t89l">38.61</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-hv44"><a href="https://huggingface.co/facebook/wav2vec2-xls-r-1b" target="_blank" rel="noopener noreferrer">facebook/wav2vec2-xls-r-1b</a></td> |
|
<td class="tg-lcf4">XL</td> |
|
<td class="tg-wssm">66.95</td> |
|
<td class="tg-wssm">75.90</td> |
|
<td class="tg-wssm">31.61</td> |
|
<td class="tg-22k7">40.41</td> |
|
<td class="tg-r0pt">62.79</td> |
|
<td class="tg-r0pt">41.99</td> |
|
<td class="tg-r0pt">43.57</td> |
|
<td class="tg-r0pt">69.79</td> |
|
<td class="tg-98fj">55.44</td> |
|
<td class="tg-98fj">99.86</td> |
|
<td class="tg-98fj">25.14</td> |
|
<td class="tg-98fj">34.58</td> |
|
</tr> |
|
<tr> |
|
<td class="tg-jqvv"><a href="https://huggingface.co/facebook/hubert-xlarge-ll60k" target="_blank" rel="noopener noreferrer">facebook/hubert-xlarge-ll60k</a></td> |
|
<td class="tg-svvj">XL</td> |
|
<td class="tg-13no">63.40</td> |
|
<td class="tg-13no">69.66</td> |
|
<td class="tg-13no">29.32</td> |
|
<td class="tg-ibhv">42.72</td> |
|
<td class="tg-heot">56.25</td> |
|
<td class="tg-heot">37.76</td> |
|
<td class="tg-heot">37.30</td> |
|
<td class="tg-heot">64.71</td> |
|
<td class="tg-imml"><span style="font-weight:700;font-style:normal"><u></span>75.69<span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-imml"><span style="font-weight:700;font-style:normal"><u></span>99.95<span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-imml"><span style="font-weight:700;font-style:normal"><u></span>47.81<span style="font-weight:700;font-style:normal"></u></span></td> |
|
<td class="tg-imml"><span style="font-weight:700;font-style:normal"><u></span>47.17<span style="font-weight:700;font-style:normal"></u></span></td> |
|
</tr> |
|
</tbody> |
|
</table> |
|
|
|
<br> |
|
<p>Best-performing model per size is highlighted in <span style="font-weight:700;font-style:normal">bold</span>. Best performing model overall is highlighted in <span style="font-weight:700;font-style:normal"><u>underlined</u></span>.</p> |
|
</body> |
|
</html> |
|
|
