TY - JOUR T1 - Speech Coding in the Brain: Representation of Vowel Formants by Midbrain Neurons Tuned to Sound Fluctuations JF - eneuro JO - eneuro DO - 10.1523/ENEURO.0004-15.2015 SP - ENEURO.0004-15.2015 AU - Laurel H. Carney AU - Tianhao Li AU - Joyce M. McDonough Y1 - 2015/07/02 UR - http://www.eneuro.org/content/early/2015/07/02/ENEURO.0004-15.2015.abstract N2 - Current models for neural coding of vowels are typically based on linear descriptions of the auditory periphery and fail at high sound levels and in background noise. These models rely on either auditory-nerve (AN) discharge rates or phase-locking to temporal fine-structure. However, both discharge rates and phase-locking saturate at moderate to high sound levels, and phase-locking is degraded in the central nervous system at mid to high frequencies. The fact that speech intelligibility is robust over a wide range of sound levels is problematic for codes that deteriorate as level increases. Additionally, a successful neural code must function for speech in background noise at levels that are tolerated by listeners. The model presented here resolves these problems and incorporates several key response properties of the nonlinear auditory periphery, including saturation, synchrony capture, and phase-locking to both fine-structure and envelope temporal features. The model also includes the properties of the auditory midbrain, where discharge rates are tuned to amplitude fluctuation rates. The nonlinear peripheral response features create contrasts in the amplitudes of low-frequency neural rate fluctuations across the population. These patterns of fluctuations result in a response profile in the midbrain that encodes vowel formants over a wide range of levels and in background noise. The hypothesized code is supported by electrophysiological recordings from the inferior colliculus of awake rabbit. This model provides information for understanding the structure of cross-linguistic vowel spaces and suggests strategies for automatic formant detection and speech enhancement for listeners with hearing loss.Significance Statement: Encoding of speech sounds is the most important function of the human auditory system. Current models for neural coding of speech fail over the range of sound levels encountered in daily life and in background noise. The acoustic structure of vowels and the properties of auditory midbrain neurons that are tuned to low-frequency amplitude fluctuations suggest a neural code for the spectral peaks (called formants) that identify vowels. The proposed neural code for speech sounds is the first that is robust over a wide range of sound levels and in background noise. These results address classic problems in auditory neuroscience and linguistics and suggest novel strategies for auditory prosthetics, automatic speech recognition, and speech enhancement for hearing aids and telephones. ER -