ThBel commited on
Commit
380622c
·
1 Parent(s): 7af9c17

Updated datsset size in README

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -48,7 +48,7 @@ The dataset is designed to support research and development of speech and dialog
48
  * **Language(s)**: English
49
  * **Modality**: Audio (speech; mono-channel; sampled at 16kHz), Text
50
  * **Interaction type**: Human conversational speech
51
- * **Utterances**: 5,332
52
  * **Speakers**: 500+
53
 
54
  Dataset splits (e.g., train/validation/test) are not predefined and may be created by downstream users as needed. Deduplication was applied to the underlying audio sources to ensure dataset splits can be made without contamination.
 
48
  * **Language(s)**: English
49
  * **Modality**: Audio (speech; mono-channel; sampled at 16kHz), Text
50
  * **Interaction type**: Human conversational speech
51
+ * **Utterances**: 7,111
52
  * **Speakers**: 500+
53
 
54
  Dataset splits (e.g., train/validation/test) are not predefined and may be created by downstream users as needed. Deduplication was applied to the underlying audio sources to ensure dataset splits can be made without contamination.