Update README.md
Browse files
README.md
CHANGED
@@ -789,17 +789,122 @@ We sugguest readers to refer to our [**Github**](https://github.com/baichuan-inc
|
|
789 |
|
790 |
<summary>click to view</summary>
|
791 |
|
792 |
-
#### Audio
|
793 |
-
|
794 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
795 |
|
796 |
-
<details>
|
797 |
|
798 |
-
|
799 |
|
800 |
-
#### Speech Generation
|
801 |
|
802 |
-
</details>
|
803 |
|
804 |
<details>
|
805 |
|
@@ -978,7 +1083,7 @@ We recommend interested scholars to visit our github repo for more details. [**G
|
|
978 |
|
979 |
|
980 |
### License
|
981 |
-
The community usage of Baichuan-Omni-1.5/Baichuan-Omni-1.5-base requires adherence to [Apache 2.0](https://github.com/baichuan-inc/Baichuan-Omni-1.5/blob/main/LICENSE) and [Community License for Baichuan-Omni-1.5 Models](https://
|
982 |
|
983 |
1. The Daily Active Users (DAU) of your or your affiliate's service or product is less than 1 million.
|
984 |
2. Neither you nor your affiliates are software service providers or cloud service providers.
|
|
|
789 |
|
790 |
<summary>click to view</summary>
|
791 |
|
792 |
+
#### Audio Comprehensive and Speech Generation
|
793 |
+
<div align="center">
|
794 |
+
<table style="margin: 0 auto; text-align: center;">
|
795 |
+
<thead>
|
796 |
+
<tr>
|
797 |
+
<th colspan="12">Audio Comprehensive Capacity</th>
|
798 |
+
</tr></thead>
|
799 |
+
<tbody>
|
800 |
+
<tr>
|
801 |
+
<td rowspan="2">Model</td>
|
802 |
+
<td rowspan="2">Size</td>
|
803 |
+
<td colspan="2">Reasoning QA</td>
|
804 |
+
<td colspan="2">Llama Questions</td>
|
805 |
+
<td colspan="2">Web Questions</td>
|
806 |
+
<td colspan="2">TriviaQA</td>
|
807 |
+
<td colspan="2">AlpacaEval</td>
|
808 |
+
</tr>
|
809 |
+
<tr>
|
810 |
+
<td>s→t</td>
|
811 |
+
<td>s→s</td>
|
812 |
+
<td>s→t</td>
|
813 |
+
<td>s→s</td>
|
814 |
+
<td>s→t</td>
|
815 |
+
<td>s→s</td>
|
816 |
+
<td>s→t</td>
|
817 |
+
<td>s→s</td>
|
818 |
+
<td>s→t</td>
|
819 |
+
<td>s→s</td>
|
820 |
+
</tr>
|
821 |
+
<tr>
|
822 |
+
<td colspan="12">Proprietary Models</td>
|
823 |
+
</tr>
|
824 |
+
<tr>
|
825 |
+
<td>GPT-4o-Audio</td>
|
826 |
+
<td>-</td>
|
827 |
+
<td><b>55.6</td>
|
828 |
+
<td>-</td>
|
829 |
+
<td><b>88.4</td>
|
830 |
+
<td>-</td>
|
831 |
+
<td><b>8.10</td>
|
832 |
+
<td>-</td>
|
833 |
+
<td><b>9.06</td>
|
834 |
+
<td>-</td>
|
835 |
+
<td><b>8.01</td>
|
836 |
+
<td>-</td>
|
837 |
+
</tr>
|
838 |
+
<tr>
|
839 |
+
<td colspan="12">Open-source Models (Pure Audio)</td>
|
840 |
+
</tr>
|
841 |
+
<tr>
|
842 |
+
<td>GLM-4-Voice</td>
|
843 |
+
<td>9B</td>
|
844 |
+
<td>-</td>
|
845 |
+
<td>26.5</td>
|
846 |
+
<td>-</td>
|
847 |
+
<td>71.0</td>
|
848 |
+
<td>-</td>
|
849 |
+
<td>5.15</td>
|
850 |
+
<td>-</td>
|
851 |
+
<td>4.66</td>
|
852 |
+
<td>-</td>
|
853 |
+
<td>4.89</td>
|
854 |
+
</tr>
|
855 |
+
<tr>
|
856 |
+
<td colspan="12">Open-source Models (Omni-modal)</td>
|
857 |
+
</tr>
|
858 |
+
<tr>
|
859 |
+
<td>VITA-1.5</td>
|
860 |
+
<td>7B</td>
|
861 |
+
<td>41.0</td>
|
862 |
+
<td>-</td>
|
863 |
+
<td>74.2</td>
|
864 |
+
<td>-</td>
|
865 |
+
<td>5.73</td>
|
866 |
+
<td>-</td>
|
867 |
+
<td>4.68</td>
|
868 |
+
<td>-</td>
|
869 |
+
<td>6.82</td>
|
870 |
+
<td>-</td>
|
871 |
+
</tr>
|
872 |
+
<tr>
|
873 |
+
<td>MiniCPM-o 2.6</td>
|
874 |
+
<td>7B</td>
|
875 |
+
<td>38.6</td>
|
876 |
+
<td>-</td>
|
877 |
+
<td>77.8</td>
|
878 |
+
<td>-</td>
|
879 |
+
<td>6.86</td>
|
880 |
+
<td>-</td>
|
881 |
+
<td>6.19</td>
|
882 |
+
<td>-</td>
|
883 |
+
<td>5.18</td>
|
884 |
+
<td>-</td>
|
885 |
+
</tr>
|
886 |
+
<tr>
|
887 |
+
<td><b>Baichuan-Omni-1.5</td>
|
888 |
+
<td>7B</td>
|
889 |
+
<td>50.0</td>
|
890 |
+
<td><b>40.9</td>
|
891 |
+
<td>78.5</td>
|
892 |
+
<td><b>75.3</td>
|
893 |
+
<td>5.91</td>
|
894 |
+
<td><b>5.52</td>
|
895 |
+
<td>5.72</td>
|
896 |
+
<td>5.31</td>
|
897 |
+
<td>7.79</td>
|
898 |
+
<td><b>6.94</td>
|
899 |
+
</tr>
|
900 |
+
</tbody>
|
901 |
+
</table>
|
902 |
+
</div>
|
903 |
|
|
|
904 |
|
905 |
+
</details>
|
906 |
|
|
|
907 |
|
|
|
908 |
|
909 |
<details>
|
910 |
|
|
|
1083 |
|
1084 |
|
1085 |
### License
|
1086 |
+
The community usage of Baichuan-Omni-1.5/Baichuan-Omni-1.5-base requires adherence to [Apache 2.0](https://github.com/baichuan-inc/Baichuan-Omni-1.5/blob/main/LICENSE) and [Community License for Baichuan-Omni-1.5 Models](https://github.com/baichuan-inc/Baichuan-Omni-1.5/blob/main/LICENSE). The Baichuan-Omni-1.5/Baichuan-Omni-1.5-base models supports commercial use. If you plan to use the Baichuan-Omni-1.5/Baichuan-Omni-1.5-base models or its derivatives for commercial purposes, please ensure that your entity meets the following conditions:
|
1087 |
|
1088 |
1. The Daily Active Users (DAU) of your or your affiliate's service or product is less than 1 million.
|
1089 |
2. Neither you nor your affiliates are software service providers or cloud service providers.
|