Dataset Viewer (First 5GB)
Search is not available for this dataset
input_features
sequencelengths 128
128
| labels
sequencelengths 4
108
|
|---|---|
[[0.4950580596923828,-0.08481216430664062,-0.16623342037200928,-0.2516758441925049,-0.48909473419189(...TRUNCATED) | [50258,50364,24466,6566,8646,2095,390,264,2117,295,264,869,3631,749,5159,32841,567,20077,670,341,445(...TRUNCATED) |
[[-0.6739928722381592,-0.5269602537155151,-0.4519672393798828,-0.41424477100372314,-0.42067980766296(...TRUNCATED) |
[
50258,
50364,
9077,
307,
257,
32890,
5513,
294,
264,
21976,
50257
] |
[[0.14082789421081543,-0.041710734367370605,0.031095802783966064,0.08194422721862793,-0.292294383049(...TRUNCATED) |
[
50258,
50364,
14076,
50257
] |
[[0.1489197015762329,-0.016188859939575195,-0.00476837158203125,-0.11558341979980469,0.0555874109268(...TRUNCATED) |
[
50258,
50364,
45,
647,
545,
5752,
12953,
281,
220,
264,
50257
] |
[[0.021378040313720703,-0.2882375717163086,-0.12592852115631104,-0.3926961421966553,-0.1653627157211(...TRUNCATED) |
[
50258,
50364,
6462,
709,
1460,
307,
1411,
294,
452,
1587,
409,
608,
304,
2114,
26338,
497,
1807,
8915,
2696,
30,
50257
] |
[[0.050065040588378906,-0.2633768320083618,-0.61018967628479,-0.2720963954925537,-0.1081223487854003(...TRUNCATED) | [50258,50364,38,5179,6837,575,4613,2272,13801,11079,13894,46331,293,1017,1176,1653,383,533,22187,13,(...TRUNCATED) |
[[-0.30163323879241943,-0.42757582664489746,-0.4647904634475708,-0.6174745559692383,-0.5075079202651(...TRUNCATED) |
[
50258,
50364,
9077,
366,
867,
955,
18361,
50257
] |
[[-0.23216915130615234,-0.3540031909942627,-0.35740959644317627,-0.4381211996078491,-0.4500160217285(...TRUNCATED) |
[
50258,
50364,
1992,
309,
307,
6869,
538,
661,
14770,
281,
1254,
264,
4991,
8587,
569,
325,
424,
50257
] |
[[0.7134931683540344,0.35570287704467773,0.31679171323776245,0.3958550691604614,0.429726243019104,0.(...TRUNCATED) |
[
50258,
50364,
9077,
362,
668,
867,
15257,
3938,
281,
652,
988,
300,
264,
561,
6909,
641,
50257
] |
[[0.09911549091339111,-0.052660226821899414,-0.25123274326324463,-0.4804532527923584,-0.135352373123(...TRUNCATED) |
[
50258,
50364,
41,
5136,
84,
11,
318,
23337,
11,
20067,
4398,
11,
16745,
20175,
11,
430,
4106,
339,
50257
] |
End of preview. Expand
in Data Studio
The input_features are nothing but the values generated after passing the dataset's audio array through a whisper processor's feature extraction and the field 'labels' consists of the tokenized(using whisper tokenizer) ground truths. The following is the link for what I did with the sarvah dataset and how I trained it on whisper-large-v3-turbo. The training steps for whisper-large-v3 are same. https://colab.research.google.com/drive/1oD0v7MWZ9WJqk7tZYThwgTUM85PTEhMN?usp=sharing
- Downloads last month
- 10