No. 180: Listen Closely: Measuring Vocal Tone in Corporate Disclosures

We examine the usefulness of machine learning approaches to measure vocal tone in corporate disclosures. We document a substantial mismatch between the widely adopted actor-based training data underlying these approaches and speech in corporate disclosures. We find that existing models achieve near-perfect vocal tone classification within their training domain. However, when tested on actual executive speech during conference calls, their performance declines to chance levels. We introduce FinVoc2Vec, a deep learning model that adapts to audio recordings of conference calls and classifies vocal tone of executive speech significantly better than chance. FinVoc2Vec estimates are associated with future firm performance, analyst reactions, and management forecast errors, and they can be used to construct profitable stock portfolios. Throughout our analyses, estimates from previous vocal tone models are largely unrelated to firm outcomes. Our findings emphasize the importance of a domain-specific approach to voice analysis in accounting and finance.