Upload README.md

README.md

@@ -19,7 +19,14 @@ pipeline_tag: audio-classification
 metrics:
 - accuracy
 ---
-This wav2vec2 based age and gender classification model is trained on [mozilla common voice](https://commonvoice.mozilla.org/).
+
+This multi-task wav2vec2 based asr model has two additional classification heads to detect:
+- age
+- gender
+... of the current speaker in one forward pass.
+
+It was trained on  [mozilla common voice](https://commonvoice.mozilla.org/).
+
 Code for training can be found [here](https://github.com/padmalcom/wav2vec2-asr-ultimate-german).
 
-*inference.py* shows, how the model can be used.
+*inference_online.py* shows, how the model can be used.