Improve model card: Add metadata, paper link, code link, and project page link
Browse filesThis PR enhances the model card for the Fun-ASR model by adding crucial metadata and improving the discoverability of related resources:
- Adds `license: apache-2.0` to the metadata, reflecting a common open-source license.
- Adds `pipeline_tag: automatic-speech-recognition` to correctly categorize the model on the Hugging Face Hub.
- Adds `library_name: funasr` to enable the automated "how to use" widget, as evidenced by the `from funasr import AutoModel` usage in the model's documentation.
- Includes prominent links to the paper ([Fun-ASR Technical Report](https://huggingface.co/papers/2509.12508)), the GitHub repository (`https://github.com/FunAudioLLM/Fun-ASR`), and the project page (`https://funaudiollm.github.io/funasr`) at the top of the model card for improved visibility.
These updates will make the model more discoverable and easier for users to understand and utilize on the Hugging Face Hub.
|
@@ -1,3 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Fun-ASR
|
| 2 |
|
| 3 |
「[简体中文](README_zh.md)」|「English」
|
|
@@ -158,13 +169,13 @@ We evaluated Fun-ASR against other state-of-the-art models on open-source benchm
|
|
| 158 |
| **Model Size** | 1.5B | 1.5B | 1.6B | - | - | - | - | 1.1B | 0.8B | 7.7B |
|
| 159 |
| **OpenSource** | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ |
|
| 160 |
| AIShell1 | 1.81 | 2.17 | 4.72 | 0.68 | 1.63 | 0.71 | 0.63 | 0.54 | 1.80 | 1.22 |
|
| 161 |
-
| AIShell2 | - | 3.47 | 4.68 | 2.27 | 2.76 | 2.86
|
| 162 |
-
| Fleurs-zh | - | 3.65 | 5.18 | 3.43 | 3.23 | 3.11
|
| 163 |
-
| Fleurs-en | 5.78 | 6.95 | 6.23 | 9.39 | 9.39 | 6.99
|
| 164 |
-
| Librispeech-clean | 2.00 | 2.17 | 1.86 | 1.58 | 2.8 | 1.32
|
| 165 |
-
| Librispeech-other | 4.19 | 4.43 | 3.43 | 2.84 | 5.69 | 2.63
|
| 166 |
-
| WenetSpeech Meeting | 6.73 | 8.21 | 18.39 | 5.69 | 7.07 | 6.24
|
| 167 |
-
| WenetSpeech Net | - | 6.33 | 11.89 | 4.66 | 4.84 | 6.45
|
| 168 |
|
| 169 |
> _Note: Seed-ASR\* results are evaluated using the official API on volcengine; GLM-ASR-nano\* results are evaluated using the open-source checkpoint._
|
| 170 |
|
|
@@ -198,4 +209,4 @@ We evaluated Fun-ASR against other state-of-the-art models on open-source benchm
|
|
| 198 |
journal={arXiv preprint arXiv:2509.12508},
|
| 199 |
year={2025}
|
| 200 |
}
|
| 201 |
-
```
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: automatic-speech-recognition
|
| 4 |
+
library_name: funasr
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
This repository contains the model presented in [Fun-ASR Technical Report](https://huggingface.co/papers/2509.12508).
|
| 8 |
+
|
| 9 |
+
Code: https://github.com/FunAudioLLM/Fun-ASR
|
| 10 |
+
Project Page: https://funaudiollm.github.io/funasr
|
| 11 |
+
|
| 12 |
# Fun-ASR
|
| 13 |
|
| 14 |
「[简体中文](README_zh.md)」|「English」
|
|
|
|
| 169 |
| **Model Size** | 1.5B | 1.5B | 1.6B | - | - | - | - | 1.1B | 0.8B | 7.7B |
|
| 170 |
| **OpenSource** | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ |
|
| 171 |
| AIShell1 | 1.81 | 2.17 | 4.72 | 0.68 | 1.63 | 0.71 | 0.63 | 0.54 | 1.80 | 1.22 |
|
| 172 |
+
| AIShell2 | - | 3.47 | 4.68 | 2.27 | 2.76 | 2.86 | 2.10 | 2.58 | 2.75 | 2.39 |
|
| 173 |
+
| Fleurs-zh | - | 3.65 | 5.18 | 3.43 | 3.23 | 3.11 | 2.68 | 4.81 | 2.56 | 2.53 |
|
| 174 |
+
| Fleurs-en | 5.78 | 6.95 | 6.23 | 9.39 | 9.39 | 6.99 | 3.03 | 10.79 | 5.96 | 4.74 |
|
| 175 |
+
| Librispeech-clean | 2.00 | 2.17 | 1.86 | 1.58 | 2.8 | 1.32 | 1.17 | 1.84 | 1.76 | 1.51 |
|
| 176 |
+
| Librispeech-other | 4.19 | 4.43 | 3.43 | 2.84 | 5.69 | 2.63 | 2.42 | 4.52 | 4.33 | 3.03 |
|
| 177 |
+
| WenetSpeech Meeting | 6.73 | 8.21 | 18.39 | 5.69 | 7.07 | 6.24 | 4.75 | 4.95 | 6.60 | 6.17 |
|
| 178 |
+
| WenetSpeech Net | - | 6.33 | 11.89 | 4.66 | 4.84 | 6.45 | 4.67 | 4.94 | 6.01 | 5.46 |
|
| 179 |
|
| 180 |
> _Note: Seed-ASR\* results are evaluated using the official API on volcengine; GLM-ASR-nano\* results are evaluated using the open-source checkpoint._
|
| 181 |
|
|
|
|
| 209 |
journal={arXiv preprint arXiv:2509.12508},
|
| 210 |
year={2025}
|
| 211 |
}
|
| 212 |
+
```
|