Update README.md
Browse files
README.md
CHANGED
@@ -29,6 +29,12 @@ size_categories:
|
|
29 |
|
30 |
- Almost the same with [Librispeech](https://huggingface.co/datasets/openslr/librispeech_asr) dataset module since i refered to the [source code](https://huggingface.co/datasets/openslr/librispeech_asr/blob/main/librispeech_asr.py).
|
31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
|
33 |
```python
|
34 |
from datasets import load_dataset
|
@@ -55,8 +61,8 @@ libripc = load_dataset("yoom618/librispeech_pc",
|
|
55 |
print(libripc)
|
56 |
|
57 |
display(Audio(libripc['train.clean.100'][0]['audio']['array'],
|
58 |
-
|
59 |
-
|
60 |
pprint(libripc['train.clean.100'][0]['audio'])
|
61 |
|
62 |
```
|
|
|
29 |
|
30 |
- Almost the same with [Librispeech](https://huggingface.co/datasets/openslr/librispeech_asr) dataset module since i refered to the [source code](https://huggingface.co/datasets/openslr/librispeech_asr/blob/main/librispeech_asr.py).
|
31 |
|
32 |
+
- three types of transcripts are given
|
33 |
+
- `text_normalized` : the trascript from Librispeech ASR
|
34 |
+
- `text`, `text_raw` : the trascripts from Librispeech-PC
|
35 |
+
|
36 |
+
|
37 |
+
|
38 |
|
39 |
```python
|
40 |
from datasets import load_dataset
|
|
|
61 |
print(libripc)
|
62 |
|
63 |
display(Audio(libripc['train.clean.100'][0]['audio']['array'],
|
64 |
+
rate=libripc['train.clean.100'][0]['audio']['sampling_rate'],
|
65 |
+
autoplay=False))
|
66 |
pprint(libripc['train.clean.100'][0]['audio'])
|
67 |
|
68 |
```
|