sanchit-gandhi HF staff commited on
Commit
18ab3aa
1 Parent(s): e3615e7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -91,7 +91,7 @@ The audio and transcriptions are in English, as per the TED talks at http://www.
91
  ### Data Splits
92
  There are three releases for the TED-LIUM corpus, progressively increasing the number of transcribed speech training data from 118 hours (Release 1), to 207 hours (Release 2), to 452 hours (Release 3).
93
 
94
- Release 1 (default config) 2012:
95
  - 774 audio talks and automatically aligned transcriptions.
96
  - Contains 118 hours of speech audio data.
97
  - Homepage: https://www.openslr.org/7/
@@ -104,7 +104,6 @@ Release 2:
104
  - Homepage: https://www.openslr.org/19/
105
 
106
  Release 3:
107
-
108
  - 2351 audio talks and automatically aligned transcriptions.
109
  - Contains 452 hours of speech audio data.
110
  - TED-LIUM 2 validation and test data: 19 TED talks with their corresponding manual transcriptions.
 
91
  ### Data Splits
92
  There are three releases for the TED-LIUM corpus, progressively increasing the number of transcribed speech training data from 118 hours (Release 1), to 207 hours (Release 2), to 452 hours (Release 3).
93
 
94
+ Release 1 (default config):
95
  - 774 audio talks and automatically aligned transcriptions.
96
  - Contains 118 hours of speech audio data.
97
  - Homepage: https://www.openslr.org/7/
 
104
  - Homepage: https://www.openslr.org/19/
105
 
106
  Release 3:
 
107
  - 2351 audio talks and automatically aligned transcriptions.
108
  - Contains 452 hours of speech audio data.
109
  - TED-LIUM 2 validation and test data: 19 TED talks with their corresponding manual transcriptions.