pere commited on
Commit
1f6a736
1 Parent(s): 2bf468c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -39
README.md CHANGED
@@ -61,22 +61,6 @@ data = load_dataset("NbAiLab/NPSC", config="16K_mp3_bokmaal", streaming=True)
61
  ### Dataset Summary
62
  The NPSC dataset contains json lines with language training data. Here is an example json line:
63
  ```json
64
- {
65
- "meeting_date": datasets.Value("string"),
66
- "sentence_order": datasets.Value("int32"),
67
- "speaker_id" : datasets.Value("int32"),
68
- "speaker_name": datasets.Value("string"),
69
- "sentence_text": datasets.Value("string"),
70
- "sentence_language_code": datasets.Value("string"),
71
- "text": datasets.Value("string"),
72
- "start_time": datasets.Value("int32"),
73
- "end_time": datasets.Value("int32"),
74
- "normsentence_text": datasets.Value("string"),
75
- "transsentence_text": datasets.Value("string"),
76
- "translated": datasets.Value("int32"),
77
- "audio": datasets.features.Audio(sampling_rate=sampling_rate),
78
-
79
- }
80
  {
81
  "sentence_id": 49853,
82
  "sentence_order": 0,
@@ -85,14 +69,12 @@ The NPSC dataset contains json lines with language training data. Here is an exa
85
  "sentence_text": "Stortingets møte er lovlig satt",
86
  "sentence_language_code": "nb-NO",
87
  "text": "Stortingets møte er lovlig satt",
88
- "start_time": 320246, "end_time": 323590,
 
89
  "normsentence_text": "Stortingets møte er lovlig satt",
90
  "transsentence_text": "Stortingets møte er lovleg sett",
91
  "translated": 1,
92
- "audio": {"path": "audio/20170110-095504_320246_323590.wav",
93
- "array": [.......]
94
- }
95
-
96
  }
97
  !!!PES: Not all the fields above and below correspond to what is actually found
98
  in the json files. It would be good if you could correct this. When you have, I can
@@ -100,25 +82,26 @@ add further descriptions of the fields where needed.!!!
100
 
101
  ```
102
  ## Data Fields (!!!!We really need a descrption here - and maybe some cleaning!!!)
103
- |**id:** | String with id to source of line and a unique identifier|
104
- |:-----------|:------------|
105
- |**sentence_order** | String with order of sentence |
106
- |**speaker id** | Integer id of speaker |
107
- | **speaker_name** | String name of speaker |
108
- | **sentence_text** | String sentence text |
109
- | **sentence_language_code** | String sentence text (!!!LIST ALL ALTERNATIVES!!!)|
110
- | **text** | String sentence text. This is a copy of "sentence_text". It is included here to make it more convenient to interleave with other datasets.|
111
- | **start_time** | int start time |
112
- | **end_time** | int end time |
113
- | **normsentence_text** | String normalised sentence text |
114
- | **transsentence_text** | String translated sentence text |
115
- | **translated** | int text translated |
116
- | **audio** | audio audio record with 'path',(mp3) 'array','sampling_rate' (48000) |
 
117
 
118
 
119
 
120
  ### Dataset Creation
121
- We are providing a **train** and a **validation** split. The standard size of the validation is a single 1GB file, while train is sharded in 1GB chunks.
122
  All files are gzipped. There is also a **test** split, which is not shared here. It is, however part of the [official release of the NPSC](https://www.nb.no/sprakbanken/ressurskatalog/oai-nb-no-sbr-58/).
123
 
124
 
@@ -139,9 +122,6 @@ The procedure for the dataset creation is described in detail in our paper.
139
  | | Male: 61.7% |
140
 
141
 
142
- !!!It would be great to know how many hours of Bokmål/Nynorsk/English(?) here!!!!!
143
- !!!PES: As long as we have the percentages, I don't think it is necessary!!!
144
-
145
  ## Considerations for Using the Data
146
  This corpus contains speech data. All recordings are of members of Parliament in a public setting, and can be distributed without any restrains.
147
 
 
61
  ### Dataset Summary
62
  The NPSC dataset contains json lines with language training data. Here is an example json line:
63
  ```json
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
64
  {
65
  "sentence_id": 49853,
66
  "sentence_order": 0,
 
69
  "sentence_text": "Stortingets møte er lovlig satt",
70
  "sentence_language_code": "nb-NO",
71
  "text": "Stortingets møte er lovlig satt",
72
+ "start_time": 320246,
73
+ "end_time": 323590,
74
  "normsentence_text": "Stortingets møte er lovlig satt",
75
  "transsentence_text": "Stortingets møte er lovleg sett",
76
  "translated": 1,
77
+ "audio": {"path": "audio/20170110-095504_320246_323590.wav","array": [.......]}
 
 
 
78
  }
79
  !!!PES: Not all the fields above and below correspond to what is actually found
80
  in the json files. It would be good if you could correct this. When you have, I can
 
82
 
83
  ```
84
  ## Data Fields (!!!!We really need a descrption here - and maybe some cleaning!!!)
85
+ |**Key** | **Type** | **Description** |
86
+ |:-----------|:------------|:------------|
87
+ |**sentence_id:** | Integer | String with id to source of line and a unique identifier|
88
+ |**sentence_order** | String | with order of sentence |
89
+ |**speaker id** | Integer | id of speaker |
90
+ | **speaker_name** | String | name of speaker |
91
+ | **sentence_text** | String | sentence text |
92
+ | **sentence_language_code** | String | sentence text (!!!LIST ALL ALTERNATIVES!!!)|
93
+ | **text** | String | sentence text. This is a copy of "sentence_text". It is included here to make it more convenient to interleave with other datasets.|
94
+ | **start_time** | Integer | start time |
95
+ | **end_time** | Integer | end time |
96
+ | **normsentence_text** | String | normalised sentence text |
97
+ | **transsentence_text** | String | translated sentence text |
98
+ | **translated** | Integer | text translated |
99
+ | **audio** | array | Encoded audio with 'path', 'array','sampling_rate' |
100
 
101
 
102
 
103
  ### Dataset Creation
104
+ We are providing a **train** and a **validation** split. The standard size of the validation is a single 1GB file, while train is charded in 1GB chunks.
105
  All files are gzipped. There is also a **test** split, which is not shared here. It is, however part of the [official release of the NPSC](https://www.nb.no/sprakbanken/ressurskatalog/oai-nb-no-sbr-58/).
106
 
107
 
 
122
  | | Male: 61.7% |
123
 
124
 
 
 
 
125
  ## Considerations for Using the Data
126
  This corpus contains speech data. All recordings are of members of Parliament in a public setting, and can be distributed without any restrains.
127