Premise
Transcript-only analysis discards signal. The linguistic tone of an earnings call has been shown to predict abnormal returns and post-earnings announcement drift, and none of that survives being reduced to text.
Language is easy to optimise. Tone is not. Management can rehearse the words; the delivery is harder to control.
Approach
The system runs FinBERT sentiment analysis over the transcript, then layers vocal stress features extracted with Wav2Vec2 and Librosa over the same timeline, producing a management confidence score.
The output of interest is the divergence between the two: where the words and the delivery disagree. On top of that the tool tracks narrative shifts across quarters and compares a company against its peers.
Result
Tone-to-text divergence scores correlated with cumulative abnormal returns across the companies tested.
Third in the UK in the CFA Institute AI Investment Challenge, against 28 university teams. The work was published by the CFA Society.