Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,34 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
metrics:
|
4 |
+
- accuracy
|
5 |
+
pipeline_tag: audio-classification
|
6 |
+
---
|
7 |
+
## 说话识别
|
8 |
+
|
9 |
+
针对通话场景中的声音如:
|
10 |
+
|
11 |
+
| sound | description |
|
12 |
+
| :------- | -------: |
|
13 |
+
| bell | 响铃 |
|
14 |
+
| music | 音乐 |
|
15 |
+
| mute | 静音(完全没有声音) |
|
16 |
+
| noise | 噪音(声音比较大的噪音) |
|
17 |
+
| noise_mute | 环境音(其实也是噪音, 但声音比较小) |
|
18 |
+
| voice | 语音(用户说话的声音, 但如果是远场说话则被认为是环境音) |
|
19 |
+
| voicemail | 语音信箱(运营商播报的语音信箱) |
|
20 |
+
| white_noise | 白噪声(一般是电话线路导致的, 嗡嗡的声音) |
|
21 |
+
|
22 |
+
些模型将以上声音区分为 "non_voice", "voice" 两种. 如下:
|
23 |
+
|
24 |
+
| sound | label |
|
25 |
+
| :------- | -------: |
|
26 |
+
| bell | non_voice |
|
27 |
+
| music | non_voice |
|
28 |
+
| mute | non_voice |
|
29 |
+
| noise | non_voice |
|
30 |
+
| noise_mute | non_voice |
|
31 |
+
| voice | voice |
|
32 |
+
| voicemail | voice |
|
33 |
+
| white_noise | voice |
|
34 |
+
|