Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,43 @@
|
|
1 |
-
---
|
2 |
-
license: cc-by-4.0
|
3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-4.0
|
3 |
+
---
|
4 |
+
|
5 |
+
# ๐ฌ ScreenTalk-XS: Sample Speech Dataset from Movies & TV ๐บ
|
6 |
+
|
7 |
+

|
8 |
+

|
9 |
+

|
10 |
+
|
11 |
+
### ๐ข **What is ScreenTalk-XS?**
|
12 |
+
**ScreenTalk-XS** is a **limited sample** from the full **ScreenTalk dataset**, containing **transcribed speech data** from selected movie and TV dialogues.
|
13 |
+
It is designed for **automatic speech recognition (ASR), natural language processing (NLP), and conversational AI research**.
|
14 |
+
|
15 |
+
โ
**This dataset serves as a free preview** of the full **ScreenTalk** dataset, which includes a much larger collection of movie and TV speech data.
|
16 |
+
|
17 |
+
---
|
18 |
+
|
19 |
+
## ๐ **Dataset Details**
|
20 |
+
- **Total Clips**: ~10% of the full dataset
|
21 |
+
- **Languages**: English
|
22 |
+
- **Content**: Transcribed movie & TV dialogues
|
23 |
+
- **Format**: `.wav` (audio) + `.tsv` (transcriptions)
|
24 |
+
- **Use Case**: Speech recognition, NLP, AI model training
|
25 |
+
|
26 |
+
๐ **The full dataset (`ScreenTalk`) contains train, test, and validation splits and is available with restricted access.**
|
27 |
+
|
28 |
+
---
|
29 |
+
|
30 |
+
## ๐ **Dataset Structure**
|
31 |
+
ScreenTalk-XS contains the following fields:
|
32 |
+
|
33 |
+
| Column | Description |
|
34 |
+
|--------|-------------|
|
35 |
+
| `audio` | Audio file path |
|
36 |
+
| `sentence` | Transcribed speech |
|
37 |
+
|
38 |
+
๐ **Example Entry**
|
39 |
+
```json
|
40 |
+
{
|
41 |
+
"audio": "path/to/audio.wav",
|
42 |
+
"sentence": "I will find you and I will train my model."
|
43 |
+
}
|