ONNX
Chinese
English
YingaoWang-casia commited on
Commit
32c2a4b
·
verified ·
1 Parent(s): 6191653

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -6
README.md CHANGED
@@ -1,6 +1,9 @@
 
 
 
1
  <div align="center">
2
 
3
- <img src="./image/Baiji_Team.png" alt="Baiji Team Logo" width="400" height="200"/>
4
 
5
  <br/>
6
 
@@ -10,9 +13,8 @@
10
 
11
  <br/>
12
 
13
- ```
14
- 47M Parameters | CPU Latency ~55ms | F1 up to 96.35% | Invalid Utterance Filtering
15
- ```
16
 
17
  <br/>
18
 
@@ -82,7 +84,7 @@
82
  Traditional approaches typically rely on a simple binary classification — "finished or not." **TurnSense goes further** by simultaneously modeling semantic completeness and invalid input detection, enabling more natural turn-taking in complex real-world scenarios and **significantly reducing false interruptions, premature responses, and noise-triggered activations**.
83
 
84
  <div align="center">
85
- <img src="./image/TurnSense.png" alt="TurnSense Three-Class Illustration" width="820"/>
86
  </div>
87
 
88
  <br/>
@@ -350,8 +352,9 @@ If you have questions or suggestions, feel free to reach out:
350
 
351
  | Channel | Contact |
352
  | :--- | :--- |
353
- | 📧 Email | huan.shen@brgroup.com · yingao.wang@brgroup.com · wei.zou@brgroup.com |
354
  | 💬 WeChat | h2538406363 |
 
355
  | 🐛 Issues | [GitHub Issues](https://github.com/Baiji-Team/TurnSense/issues) |
356
  | 🔀 PR | [Pull Requests](https://github.com/Baiji-Team/TurnSense/pulls) |
357
 
 
1
+ ---
2
+ license: apache-2.0
3
+ ---
4
  <div align="center">
5
 
6
+ <img src="./TenSense/image/Baiji_Team.png" alt="Baiji Team Logo" width="1000" height="450"/>
7
 
8
  <br/>
9
 
 
13
 
14
  <br/>
15
 
16
+ <center><strong>47M 参数 | CPU 延迟 ~55ms | F1 高达 96.35% | 无效语义过滤</strong></center>
17
+
 
18
 
19
  <br/>
20
 
 
84
  Traditional approaches typically rely on a simple binary classification — "finished or not." **TurnSense goes further** by simultaneously modeling semantic completeness and invalid input detection, enabling more natural turn-taking in complex real-world scenarios and **significantly reducing false interruptions, premature responses, and noise-triggered activations**.
85
 
86
  <div align="center">
87
+ <img src="./TenSense/image/TurnSense.svg" alt="TurnSense Three-Class Illustration" width="820"/>
88
  </div>
89
 
90
  <br/>
 
352
 
353
  | Channel | Contact |
354
  | :--- | :--- |
355
+ | 📧 Email | [huan.shen@brgroup.com](mailto:huan.shen@brgroup.com) · [yingao.wang@brgroup.com](mailto:yingao.wang@brgroup.com) · [wei.zou@brgroup.com](mailto:wei.zou@brgroup.com) |
356
  | 💬 WeChat | h2538406363 |
357
+ | 👥 WeChat Group | Scan the QR code to join the group<br><img src="TenSense/image/wechat.jpg" alt="WeChat group QR code" width="220" /> |
358
  | 🐛 Issues | [GitHub Issues](https://github.com/Baiji-Team/TurnSense/issues) |
359
  | 🔀 PR | [Pull Requests](https://github.com/Baiji-Team/TurnSense/pulls) |
360