Datasets:
audio dict | text stringclasses 10
values | duration_s float32 6.13 14.1 | snr_db float32 21 41.3 | wps float32 0.09 0.89 | original_text stringclasses 10
values | split stringclasses 1
value |
|---|---|---|---|---|---|---|
{"bytes":"UklGRsTxBABXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | "今年正月份 ê 總統大選，台灣人民拍開歷史 ê 新頁，連紲第三遍將國家 ê (...TRUNCATED) | 10.1239 | 30 | 0.89 | "今年正月份 ê 總統大選，台灣人民拍開歷史 ê 新頁，連紲第三遍將國家 ê (...TRUNCATED) | train |
{"bytes":"UklGRvR0BABXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | 一方面突破戒嚴，拍破黨禁、報禁，追求百分之百 ê 言論自由。 | 9.1254 | 25.6 | 0.33 | 一方面突破戒嚴，拍破黨禁、報禁，追求百分之百 ê 言論自由。 | train |
{"bytes":"UklGRoj+AgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | 秉持清廉、勤政、愛鄉土 ê 精神，和人民徛做伙。 | 6.1301 | 21 | 0.49 | 秉持清廉、勤政、愛鄉土 ê 精神，和人民徛做伙。 | train |
{"bytes":"UklGRv7kBgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | "就是期待咱會當繼續帶領台灣，行對 ê 路，深化民主憲政，全力鞏固和平 k(...TRUNCATED) | 14.1177 | 26.6 | 0.35 | "就是期待咱會當繼續帶領台灣，行對 ê 路，深化民主憲政，全力鞏固和平 k(...TRUNCATED) | train |
{"bytes":"UklGRlh7AwBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | 這个過程中，民進黨 mā 由在野到執政， | 7.1286 | 23.299999 | 0.42 | 這个過程中，民進黨 mā 由在野到執政， | train |
{"bytes":"UklGRoj+AgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | 由執政淪為在野，又 koh 再一次執政。 | 6.1301 | 27.1 | 0.49 | 由執政淪為在野，又 koh 再一次執政。 | train |
{"bytes":"UklGRv7kBgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | "所以今仔日，咱有機會 tī 遮召開全國黨員代表大會，咱應該愛用感謝先輩(...TRUNCATED) | 14.1177 | 31.799999 | 0.64 | "所以今仔日，咱有機會 tī 遮召開全國黨員代表大會，咱應該愛用感謝先輩(...TRUNCATED) | train |
{"bytes":"UklGRpJuBQBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | "另外一方面推動國會全面改選、總統直選，帶動國家民主進步、社會多元發(...TRUNCATED) | 11.1224 | 22.799999 | 0.09 | "另外一方面推動國會全面改選、總統直選，帶動國家民主進步、社會多元發(...TRUNCATED) | train |
{"bytes":"UklGRoj+AgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | （現場 pho̍k-á-siaⁿ） | 6.1301 | 34.5 | 0.33 | （現場 pho̍k-á-siaⁿ） | train |
{"bytes":"UklGRoj+AgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAATElTVBoAAABJTkZPSVNGVA0AAABMYXZmNjAuMy4xMD(...TRUNCATED) | 咱民進黨，誕生 tī 專制獨裁年代， | 6.1301 | 41.299999 | 0.49 | 咱民進黨，誕生 tī 專制獨裁年代， | train |
lai-ching-te-speech
ASR dataset packaged from SRT-aligned audio clips.
Dataset info
| Split | Samples |
|---|---|
| train | 10 |
| validation | 0 |
| test | 0 |
Total audio: 0.0 hours
Usage
from datasets import load_dataset, Audio
ds = load_dataset("lai-ching-te-speech")
ds = ds.cast_column("audio", Audio(sampling_rate=16000))
# Access a sample
sample = ds["train"][0]
print(sample["text"]) # transcript
print(sample["audio"]) # {"array": ..., "sampling_rate": ...}
Columns
audio— raw WAV bytes embedded in Parquettext— ASR-normalized transcript (lowercase, no punctuation)duration_s— clip duration in secondssnr_db— estimated signal-to-noise ratiowps— words per second (speaking rate)original_text— raw subtitle text before normalization
- Downloads last month
- 26