Datasets:
id stringclasses 5
values | lang stringclasses 5
values | norm stringclasses 2
values | license null | source null | redistributable bool 1
class | n_tokens int64 27 171 | word_id listlengths 27 171 | start listlengths 27 171 | end listlengths 27 171 | line_break listlengths 27 171 | cue_break listlengths 27 171 | text stringclasses 5
values | surface listlengths 27 171 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
en_netflix | en | netflix | null | null | true | 41 | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40
] | [
0.5,
0.562,
0.812,
0.938,
1.25,
1.625,
1.812,
2,
2.438,
2.688,
3.3,
3.423,
3.732,
3.979,
4.102,
4.287,
4.596,
4.904,
5.089,
5.398,
5.645,
6.6,
7.01,
7.283,
7.488,
7.966,
8.102,
8.239,
8.444,
8.649,
8.99,
9.8,
9.998,
10.261,
10.393,
10.92,
11.11... | [
0.562,
0.812,
0.938,
1.25,
1.625,
1.812,
2,
2.438,
2.688,
3,
3.423,
3.732,
3.979,
4.102,
4.287,
4.596,
4.904,
5.089,
5.398,
5.645,
6.2,
7.01,
7.283,
7.488,
7.966,
8.102,
8.239,
8.444,
8.649,
8.99,
9.4,
9.998,
10.261,
10.393,
10.92,
11.117,
11.2... | [
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0
] | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1
] | I woke up early today, and the morning felt calm. We drove down to the coast where the waves were enormous. Nobody else was around, so we had the whole beach. She said it reminded her of the summers back home. | [
"I",
"woke",
"up",
"early",
"today,",
"and",
"the",
"morning",
"felt",
"calm.",
"We",
"drove",
"down",
"to",
"the",
"coast",
"where",
"the",
"waves",
"were",
"enormous.",
"Nobody",
"else",
"was",
"around,",
"so",
"we",
"had",
"the",
"whole",
"beach.",
"S... |
de_netflix | de | netflix | null | null | true | 27 | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26
] | [
0.4,
0.545,
1.752,
2.331,
2.621,
3.5,
3.647,
3.893,
4.041,
5.073,
5.417,
5.81,
6.8,
6.971,
7.314,
7.543,
7.714,
8.171,
8.629,
9.371,
10,
10.347,
10.578,
10.751,
11.156,
11.329,
11.733
] | [
0.545,
1.752,
2.331,
2.621,
3.2,
3.647,
3.893,
4.041,
5.073,
5.417,
5.81,
6.4,
6.971,
7.314,
7.543,
7.714,
8.171,
8.629,
9.371,
9.6,
10.347,
10.578,
10.751,
11.156,
11.329,
11.733,
12.6
] | [
0,
1,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0
] | [
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
1
] | Der Geschwindigkeitsbegrenzer funktioniert wieder einwandfrei. Wir haben die Donaudampfschifffahrt gestern zusammen ausprobiert. Sie sagte, dass die Aussicht wirklich atemberaubend war. Danach sind wir langsam zum Bahnhof zurückgegangen. | [
"Der",
"Geschwindigkeitsbegrenzer",
"funktioniert",
"wieder",
"einwandfrei.",
"Wir",
"haben",
"die",
"Donaudampfschifffahrt",
"gestern",
"zusammen",
"ausprobiert.",
"Sie",
"sagte,",
"dass",
"die",
"Aussicht",
"wirklich",
"atemberaubend",
"war.",
"Danach",
"sind",
"wir",
... |
ja_netflix | ja | netflix | null | null | true | 98 | [
0,
0,
1,
2,
2,
2,
3,
3,
4,
4,
5,
6,
6,
7,
7,
8,
9,
10,
10,
11,
11,
12,
13,
14,
15,
16,
16,
16,
17,
17,
17,
18,
19,
19,
19,
20,
21,
22,
23,
24,
25,
25,
26,
27,
27,
28,
29,
30,
30,
31,
32,
32,
33,
33,
34,
35,... | [
0.5,
0.609,
0.717,
0.826,
0.935,
1.043,
1.152,
1.261,
1.37,
1.478,
1.587,
1.696,
1.804,
1.913,
2.022,
2.13,
2.239,
2.348,
2.457,
2.565,
2.674,
2.783,
2.891,
3.3,
3.421,
3.542,
3.662,
3.783,
3.904,
4.025,
4.146,
4.267,
4.388,
4.508,
4.629,
4.75,
... | [
0.609,
0.717,
0.826,
0.935,
1.043,
1.152,
1.261,
1.37,
1.478,
1.587,
1.696,
1.804,
1.913,
2.022,
2.13,
2.239,
2.348,
2.457,
2.565,
2.674,
2.783,
2.891,
3,
3.421,
3.542,
3.662,
3.783,
3.904,
4.025,
4.146,
4.267,
4.388,
4.508,
4.629,
4.75,
4.871,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | 今朝はとても早く起きて海岸まで車で行きました。波がとても大きくてまわりには誰もいませんでした。彼女はこの景色を見ると故郷の夏を思い出すと言いました。それから私たちはゆっくり駅まで歩いて戻りました。 | [
"今",
"朝",
"は",
"と",
"て",
"も",
"早",
"く",
"起",
"き",
"て",
"海",
"岸",
"ま",
"で",
"車",
"で",
"行",
"き",
"ま",
"し",
"た",
"。",
"波",
"が",
"と",
"て",
"も",
"大",
"き",
"く",
"て",
"ま",
"わ",
"り",
"に",
"は",
"誰",
"も",
"い",
"ま",
"せ",
"ん",
"で",
"し",
"... |
zh_bbc | zh | bbc | null | null | true | 66 | [
0,
0,
1,
1,
2,
3,
4,
4,
5,
5,
6,
7,
7,
8,
8,
9,
10,
11,
11,
12,
13,
13,
14,
14,
15,
16,
16,
17,
17,
18,
19,
20,
20,
21,
22,
23,
24,
24,
25,
25,
26,
27,
27,
28,
29,
30,
30,
31,
31,
32,
33,
33,
34,
35,
35,
36... | [
0.4,
0.52,
0.64,
0.76,
0.88,
1,
1.12,
1.24,
1.36,
1.48,
1.6,
1.72,
1.84,
1.96,
2.08,
2.2,
2.32,
2.44,
2.56,
2.68,
3,
3.186,
3.371,
3.557,
3.743,
3.929,
4.114,
4.3,
4.486,
4.671,
4.857,
5.043,
5.229,
5.414,
5.9,
6.032,
6.163,
6.295,
6.426,
... | [
0.52,
0.64,
0.76,
0.88,
1,
1.12,
1.24,
1.36,
1.48,
1.6,
1.72,
1.84,
1.96,
2.08,
2.2,
2.32,
2.44,
2.56,
2.68,
2.8,
3.186,
3.371,
3.557,
3.743,
3.929,
4.114,
4.3,
4.486,
4.671,
4.857,
5.043,
5.229,
5.414,
5.6,
6.032,
6.163,
6.295,
6.426,
6.55... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | 今天早上我很早就起来了然后开车去了海边。海浪非常大周围一个人也没有。她说看到这样的风景就会想起家乡的夏天。后来我们慢慢地走回了车站。 | [
"今",
"天",
"早",
"上",
"我",
"很",
"早",
"就",
"起",
"来",
"了",
"然",
"后",
"开",
"车",
"去",
"了",
"海",
"边",
"。",
"海",
"浪",
"非",
"常",
"大",
"周",
"围",
"一",
"个",
"人",
"也",
"没",
"有",
"。",
"她",
"说",
"看",
"到",
"这",
"样",
"的",
"风",
"景",
"就",
"会",
"... |
th_netflix | th | netflix | null | null | true | 171 | [
0,
0,
0,
0,
0,
0,
0,
1,
1,
1,
2,
2,
2,
2,
3,
3,
3,
3,
3,
3,
3,
4,
4,
4,
5,
5,
5,
5,
6,
6,
6,
6,
6,
7,
7,
8,
8,
8,
9,
9,
9,
9,
9,
9,
10,
10,
10,
10,
10,
11,
11,
11,
12,
12,
12,
13,
13,
13,
13,
13,
... | [
0.5,
0.561,
0.623,
0.684,
0.745,
0.807,
0.868,
0.93,
0.991,
1.052,
1.114,
1.175,
1.236,
1.298,
1.359,
1.42,
1.482,
1.543,
1.605,
1.666,
1.727,
1.789,
1.85,
1.911,
1.973,
2.034,
2.095,
2.157,
2.218,
2.28,
2.341,
2.402,
2.464,
2.525,
2.586,
2.648,
... | [
0.561,
0.623,
0.684,
0.745,
0.807,
0.868,
0.93,
0.991,
1.052,
1.114,
1.175,
1.236,
1.298,
1.359,
1.42,
1.482,
1.543,
1.605,
1.666,
1.727,
1.789,
1.85,
1.911,
1.973,
2.034,
2.095,
2.157,
2.218,
2.28,
2.341,
2.402,
2.464,
2.525,
2.586,
2.648,
2.709... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | เช้านี้ฉันตื่นแต่เช้ามากแล้วขับรถไปที่ชายหาดคลื่นสูงมากจริงๆและรอบตัวไม่มีใครเลยเธอบอกว่าเห็นวิวแบบนี้ก็นึกถึงฤดูร้อนที่บ้านเกิดหลังจากนั้นเราก็เดินกลับไปที่สถานีอย่างช้าๆ | [
"เ",
"ช",
"้",
"า",
"น",
"ี",
"้",
"ฉ",
"ั",
"น",
"ต",
"ื",
"่",
"น",
"แ",
"ต",
"่",
"เ",
"ช",
"้",
"า",
"ม",
"า",
"ก",
"แ",
"ล",
"้",
"ว",
"ข",
"ั",
"บ",
"ร",
"ถ",
"ไ",
"ป",
"ท",
"ี",
"่",
"ช",
"า",
"ย",
"ห",
"า",
"ด",
"ค",
"... |
Cue subtitle break dataset
Pairs of continuous transcript + timing and the gold break positions a
professional placed, obtained by inverting professional SRT/VTT (removing the
breaks; see cue/invert.py). Used to train the BreakScorer and to measure the
upper bound of re-segmentation quality (SubER).
Fields (one row per clip)
| field | notes |
|---|---|
id, lang, norm |
clip id, language, source normative profile (netflix/bbc/…) |
license, source |
per-clip licence and provenance |
redistributable |
whether the transcript text may be redistributed |
n_tokens, word_id, start, end |
token count, word grouping, per-token timing (s) |
line_break, cue_break |
gold labels: 1 if a line/cue break follows that token |
text, surface |
present only when redistributable=true |
surface_len |
present instead of surface for annotation-only rows |
Licensing & redistribution
Each clip's licence is confirmed and recorded individually. Material that may
not be redistributed is included in annotation-only form: timing and break
positions are kept, the transcript body (text/surface) is dropped. No ASR or
translation is performed; inputs are transcripts + timing only. Splits are
labelled so downstream users know which are annotation-only.
Build
python scripts/build_dataset.py --manifest <manifest.jsonl> --push-to <org>/cue-breaks
- Downloads last month
- -