ASLP-lab commited on
Commit
0f0596e
·
verified ·
1 Parent(s): a5948a7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -5
README.md CHANGED
@@ -3,12 +3,15 @@ license: cc-by-nc-4.0
3
  ---
4
  # WSYue-ASR-eval: Cantonese ASR Benchmark
5
 
6
- To address the unique linguistic characteristics of Cantonese in speech recognition, we propose **WSYue-ASR-eval**, a benchmark specifically designed for evaluating Cantonese ASR systems. It is tailored to assess model performance across diverse lengths, domains, and linguistic phenomena of Cantonese speech.
7
- To achieve this, the benchmark provides the following key features:
8
- - Annotated through multiple rounds of manual labeling
9
- - Includes rich tags such as text transcription, emotion, age, and gender
 
 
 
10
  - Covers Cantonese-English code-switching and multi-domain conditions
11
- - Enables comprehensive evaluation across varying speech lengths
12
 
13
  ## WSYue-ASR-eval Subsets
14
  | Set | Duration | Speakers | Hours |
 
3
  ---
4
  # WSYue-ASR-eval: Cantonese ASR Benchmark
5
 
6
+ To address the unique linguistic characteristics of Cantonese in speech recognition, we propose **WSYue-ASR-eval**, a benchmark specifically designed for evaluating Cantonese ASR systems. It is tailored to assess model performance across diverse lengths, domains, and linguistic phenomena of Cantonese speech.
7
+
8
+ The test set annotations are provided by Beijing AISHELL Technology Co., Ltd.
9
+
10
+ Key features:
11
+ - Annotated through multiple rounds of manual labeling
12
+ - Includes rich tags such as text transcription, emotion, age, and gender
13
  - Covers Cantonese-English code-switching and multi-domain conditions
14
+ - Enables comprehensive evaluation across varying speech lengths
15
 
16
  ## WSYue-ASR-eval Subsets
17
  | Set | Duration | Speakers | Hours |