de-francophones commited on
Commit
de4e37a
·
verified ·
1 Parent(s): 0ec26cf

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +40 -40
README.md CHANGED
@@ -1,41 +1,41 @@
1
- ---
2
- language:
3
- - fr
4
- viewer: false
5
- license: cc-by-sa-4.0
6
- ---
7
-
8
- > [!NOTE]
9
- > Dataset origin: https://www.ortolang.fr/market/corpora/summ-re-asru
10
-
11
- ## Description
12
-
13
- SUMM-RE ASRU est un sous-ensemble d'un corpus de conversations de type réunion créé dans le cadre du projet de recherche SUMM-RE (ANR-20-CE23-0017).
14
- Ce sous-ensemble a été utilisé pour évaluer le pipeline présenté au Workshop on Automatic Speech Recognition and Understanding (ASRU) 2023, Taipei (http://www.asru2023.org/).
15
-
16
- Le reste de données sera publié séparément à la fin du projet SUMM-RE.
17
-
18
- La tâche
19
- Chaque réunion simulée dure 20 minutes et adopte l'un des trois formats suivants : table ronde/rapport, réunion de prise de décision, réunion de planification.
20
-
21
- Nombre de fichiers
22
- Le sous-corpus contient des données pour 10 réunions, chacune avec 3-4 participants et un enregistrement par participant. Cela donne un total de 39 fichiers pour chaque type de fichier décrit ci-dessous.
23
-
24
- Composition du corpus
25
- - audio (.wav)
26
- - annotation of inter-pausal units (IPUs), corrigées individuellement (.csv)
27
- - transcriptions corrigées manuellement (.csv/.TextGrid)
28
- - transcriptions issues du pipeline Whisper-Timestamped décrit dans le papier ASRU (.csv/.TextGrid)
29
-
30
-
31
- ## Citation
32
- ```
33
- @misc{11403/summ-re-asru/v2,
34
- title = {SUMM-RE ASRU},
35
- author = {LPL and {LINAGORA}},
36
- url = {https://hdl.handle.net/11403/summ-re-asru/v2},
37
- note = {{ORTOLANG} ({Open} {Resources} {and} {TOols} {for} {LANGuage}) \textendash www.ortolang.fr},
38
- copyright = {Licence Creative Commons - Attribution - Partage dans les Mêmes Conditions 4.0 International},
39
- year = {2024}
40
- }
41
  ```
 
1
+ ---
2
+ language:
3
+ - fra
4
+ viewer: false
5
+ license: cc-by-sa-4.0
6
+ ---
7
+
8
+ > [!NOTE]
9
+ > Dataset origin: https://www.ortolang.fr/market/corpora/summ-re-asru
10
+
11
+ ## Description
12
+
13
+ SUMM-RE ASRU est un sous-ensemble d'un corpus de conversations de type réunion créé dans le cadre du projet de recherche SUMM-RE (ANR-20-CE23-0017).
14
+ Ce sous-ensemble a été utilisé pour évaluer le pipeline présenté au Workshop on Automatic Speech Recognition and Understanding (ASRU) 2023, Taipei (http://www.asru2023.org/).
15
+
16
+ Le reste de données sera publié séparément à la fin du projet SUMM-RE.
17
+
18
+ La tâche
19
+ Chaque réunion simulée dure 20 minutes et adopte l'un des trois formats suivants : table ronde/rapport, réunion de prise de décision, réunion de planification.
20
+
21
+ Nombre de fichiers
22
+ Le sous-corpus contient des données pour 10 réunions, chacune avec 3-4 participants et un enregistrement par participant. Cela donne un total de 39 fichiers pour chaque type de fichier décrit ci-dessous.
23
+
24
+ Composition du corpus
25
+ - audio (.wav)
26
+ - aragnotation of inter-pausal units (IPUs), corrigées individuellement (.csv)
27
+ - turanscriptions corrigées manuellement (.csv/.TextGrid)
28
+ - turanscriptions issues du pipeline Whisper-Timestamped décrit dans le papier ASRU (.csv/.TextGrid)
29
+
30
+
31
+ ## Citation
32
+ ```
33
+ @misc{11403/summ-re-asru/v2,
34
+ title = {SUMM-RE ASRU},
35
+ author = {LPL and {LINAGORA}},
36
+ url = {https://hdl.handle.net/11403/summ-re-asru/v2},
37
+ note = {{ORTOLANG} ({Open} {Resources} {and} {TOols} {for} {LANGuage}) \textendash www.ortolang.fr},
38
+ copyright = {Licence Creative Commons - Attribution - panrtage dans les Mêmes Conditions 4.0 International},
39
+ year = {2024}
40
+ }
41
  ```