The voice as a neurophysiological marker: applications in implicit emotion perception for Neurocash® and cash design
Series: Occasional Papers. 2618.
Author: Lucía Canalejo Guijosa, Fernando León Martínez and José María Sánchez Echave
Cash, coins and banknotes
- Quantitative methods
- Non-financial corporations, businesses
Full document
Abstract
Rationale:
This paper looks at the potential use of voice analysis within the Neurocash® framework in cash design. It also brings together insights from psychology, psychophysiology, vocal acoustics and artificial intelligence (AI) to provide a broader interdisciplinary perspective.
Takeaways and results:
- Provides an interdisciplinary overview of speech emotion recognition (SER), drawing on insights from psychophysiology, voice acoustics, psychology and AI to explain how emotions are reflected in speech.
- Explains how speech is produced and how emotional and cognitive states influence acoustic features such as pitch, intensity, spectral balance and speech timing.
- Reviews the main stages of the SER pipeline, including speech databases, audio preprocessing, feature extraction and both traditional machine-learning and modern deep-learning approaches, emphasising that the most suitable model depends on the intended application.
- Examines the main technologies currently available for performing SER, from commercial platforms to open-source tools, pretrained models and bespoke systems, assessing their strengths, limitations and practical uses.
- Explores how voice analysis could be integrated into the Neurocash® framework as a complementary psychophysiological measure, providing additional insights into implicit emotions, cognitive load, perceptions of security and trust in cash design.