Update README.md

README.md

@@ -80,10 +80,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Whisper Medium ID - FLEURS-CV-LBV - Augmented
 
-This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the:
+This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the following datasets:
 - [mozilla-foundation/common_voice_11_0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0)
 - [google/fleurs](https://huggingface.co/datasets/google/fleurs)
-- [indonesian-nlp/librivox-indonesia](https://huggingface.co/datasets/indonesian-nlp/librivox-indonesia) datasets.
+- [indonesian-nlp/librivox-indonesia](https://huggingface.co/datasets/indonesian-nlp/librivox-indonesia)
 
 It achieves the following results on the evaluation set (Common Voice 11.0):
 - Loss: 0.2788
@@ -100,10 +100,20 @@ More information needed
 
 ## Training and evaluation data
 
-More information needed
+Training:
+- [mozilla-foundation/common_voice_11_0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0) (train+validation)
+- [google/fleurs](https://huggingface.co/datasets/google/fleurs) (train+validation)
+- [indonesian-nlp/librivox-indonesia](https://huggingface.co/datasets/indonesian-nlp/librivox-indonesia) (train)
+
+Evaluation:
+- [mozilla-foundation/common_voice_11_0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0) (test)
+- [google/fleurs](https://huggingface.co/datasets/google/fleurs) (test)
+- [indonesian-nlp/librivox-indonesia](https://huggingface.co/datasets/indonesian-nlp/librivox-indonesia) (test)
 
 ## Training procedure
 
+Datasets were augmented on-the-fly using [audiomentations](https://github.com/iver56/audiomentations) via PitchShift, AddGaussianNoise and TimeStretch transformations at `p=0.3`.
+
 ### Training hyperparameters
 
 The following hyperparameters were used during training: