Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
dataset_info:
|
| 3 |
+
features:
|
| 4 |
+
- name: audio
|
| 5 |
+
dtype: audio
|
| 6 |
+
- name: text
|
| 7 |
+
dtype: string
|
| 8 |
+
- name: speaker_id
|
| 9 |
+
dtype: string
|
| 10 |
+
splits:
|
| 11 |
+
- name: train
|
| 12 |
+
num_examples: 10000
|
| 13 |
+
configs:
|
| 14 |
+
- config_name: default
|
| 15 |
+
data_files:
|
| 16 |
+
- split: train
|
| 17 |
+
path: data/train-*.parquet
|
| 18 |
+
---
|
| 19 |
+
|
| 20 |
+
# Synthetic Data JA
|
| 21 |
+
|
| 22 |
+
10K synthetic Japanese conversational TTS audio samples.
|
| 23 |
+
|
| 24 |
+
- **Speakers**: 10,000 unique speakers
|
| 25 |
+
- **Total samples**: 10,000
|
| 26 |
+
- **Sample rate**: 48 kHz
|
| 27 |
+
- **Format**: WAV embedded in Parquet
|
| 28 |
+
- **Language**: Japanese
|