Update README.md
Browse files
README.md
CHANGED
|
@@ -55,8 +55,6 @@ voxpopuli_croatian = load_dataset("Edmon02/hyvoxpopuli")
|
|
| 55 |
|
| 56 |
* automatic-speech-recognition: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER).
|
| 57 |
|
| 58 |
-
Accented English subset can also be used for research in ASR for accented speech (15 L2 accents)
|
| 59 |
-
|
| 60 |
## Dataset Structure
|
| 61 |
|
| 62 |
### Data Instances
|
|
|
|
| 55 |
|
| 56 |
* automatic-speech-recognition: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER).
|
| 57 |
|
|
|
|
|
|
|
| 58 |
## Dataset Structure
|
| 59 |
|
| 60 |
### Data Instances
|