Airwave is an audio profiling tool that predicts the age, gender and height of a speaker based on their voice characteristics in an audio recording.

It fine-tunes the Natural Language Processing and open-sourced Wav2Vec2 speech model.

This project is currently in testing, with further features in development:

• Accent Detection – dominant accent classification of an English language speaker.

• Audio Tempering Detection – Identification of where a portion of audio recording has been tampered with or synthetically generated.

• Audio Transcription – Key phrase recognition to trigger automated AI transcription of live or recorded audio files.