Update README.md
Browse files
README.md
CHANGED
@@ -19,3 +19,27 @@ In our original work, we directly transcribed the speech with Whisper Medium. Ho
|
|
19 |
|
20 |
## Intended Use
|
21 |
This set of data can be used for Accented ASR research!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
|
20 |
## Intended Use
|
21 |
This set of data can be used for Accented ASR research!
|
22 |
+
|
23 |
+
## Code Example
|
24 |
+
|
25 |
+
```python
|
26 |
+
from datasets import load_dataset
|
27 |
+
dataset = load_dataset("sylviali/EDEN_ASR_Data", split="train")
|
28 |
+
|
29 |
+
print(dataset)
|
30 |
+
|
31 |
+
# Download the audio to a local file
|
32 |
+
import urllib
|
33 |
+
urllib.request.urlretrieve(dataset[0]["audio_url"], "audio.wav")
|
34 |
+
|
35 |
+
# Extract the ASR transcript
|
36 |
+
print(dataset[0]["corrected_whisper_transcript"])
|
37 |
+
|
38 |
+
# Extract the emotion label
|
39 |
+
print(dataset[0]["emotion_label"])
|
40 |
+
|
41 |
+
# Check negative emotion audio clips
|
42 |
+
negative_emotion_clips = dataset.filter(lambda example: example["emotion_label"] == "Negative")
|
43 |
+
print(len(negative_emotion_clips))
|
44 |
+
print(negative_emotion_clips[0])
|
45 |
+
```
|