ehcalabres
/
wav2vec2-lg-xlsr-en-speech-emotion-recognition
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update README.md

This commit is contained in:
Enrique Hernández Calabrés 2021-07-14 12:47:57 +00:00 committed by huggingface-web
parent ebe7999a05
commit 874fe6191e
1 changed files with 10 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
README.md
View File

@ -11,9 +11,13 @@ model_index:
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-lg-xlsr-en-speech-emotion-recognition
# Speech Emotion Recognition By Fine-Tuning Wav2Vec 2.0
This model is a fine-tuned version of [jonatasgrosman/wav2vec2-large-xlsr-53-english](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-english) for a Speech Emotion Recognition (SER) task.
__Important:__ This model it's not yet implementable due to missing built-in functions in HuggingFace for speech classification tasks. I'm working on the instructions of how to use it and a repository where the code will be available soon. Thank you anyway!
The model is a fine-tuned version of [jonatasgrosman/wav2vec2-large-xlsr-53-english](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-english) for a Speech Emotion Recognition (SER) task.
The dataset used to fine-tune the original pre-trained model is the [RAVDESS dataset](https://zenodo.org/record/1188976#.YO6yI-gzaUk). This dataset provides 1440 samples of recordings from actors performing on 8 different emotions in English, which are:
@ -72,6 +76,10 @@ The following hyperparameters were used during training:
| 0.4581 | 2.72 | 390 | 0.4719 | 0.8467 |
| 0.3967 | 2.93 | 420 | 0.5023 | 0.8223 |
## Contact
Any doubt, contact me on [Twitter](https://twitter.com/ehcalabres) (GitHub repo soon).
### Framework versions