🛠 Contribute: Found a typo? Or any other change in the description that you would like to see? Please consider sending us a pull request in the public repo of the challenge here.

🕵️ Introduction

We humans rely on our community's feedback and review for so many things. When our friends tell us about their visit to the new restaurant, we can gauge whether they had a positive or a negative experience. When our family talks about the new movie, we can know whether they enjoyed it or not. But do you think machines can identify sentiment based on the sound clips of reviews?

This challenge merges multiple domains of AI to address an interesting question. Can you build a model that can identify sentiment from sound clips?
Understand with code! Here is getting started code for you.😄

💾 Dataset

The dataset contains the audio file containing reviews by the patients. The dataset is divided into train and validation set with 15000 and 2000 sample in them respectively. The label for each audio is present in train.csv and val.csv corresponding to their id.

📁 Files

Following files are available in the resources section:

• train.csv : (15000 samples) This training CSV file contains the column wav_id which corresponds with the audio id in train.zip and label the column containing if the voice sentiment is positive negative or neutral. 0 means positive, 1 means neutral, and 2 means negative

• val.csv : (2000 samples) This validation CSV file contains the column wav_id which corresponds with the audio id in val.zip and label the column containing if the voice sentiment is positive negative or neutral. 0 means positive, 1 means neutral, and 2 means negative.

• train.zip : (15000 samples) This training zip file contains .wav files containing the project reviews sound.

• val.zip : (2000 samples) This validation zip file contains .wav files containing the project reviews sound.

• test.zip : (7000 samples) This testing zip file contains .wav files containing the project review sound.

🚀 Submission

• Prepare a CSV containing [wavid, label] as headers, where wav_id is the audio id and label is a digit in the range [0-2] denoting positive, neutral, and negative reviews respectively.
• Sample submission format available at sample_submission.csv in the resources section.

Make your first submission here 🚀 !!

🖊 Evaluation Criteria

During the evaluation, F1 score will be used to test the efficiency of the model where,

🔗 Links

• 💪 Challenge Page: https://www.aicrowd.com/challenges/sousen
• 🗣️ Discussion Forum: https://www.aicrowd.com/challenges/sousen/discussion
• 🏆 Leaderboard: https://www.aicrowd.com/challenges/sousen/leaderboards

📱 Contact

• Shubhamai