This study examined English emotional prosody and identified differences in vocal expression of love and sorrow among first language (L1) speakers, learners of English as a foreign language (EFL), and text-to-speech systems (TTS). American professional narrators (AE) and Japanese college students (JP) were instructed to read aloud and record a love letter and a condolence letter. The same letters were also read by synthetic speakers on a web-based TTS. The measurements targeted "booster expressions"-lexical grammatical items of various types which increase the effectiveness of emotional utterances (Quirk & Greenbaum, 1973). The results showed that onset consonant duration, pitch range, intensity, and speech rate had significant differences among groups. These findings suggest that all these features contribute significantly to how AE express love and sorrow, which is still not available through the current default system of TTS. The findings have an impact on the broader topic of speaking English as a foreign language inside and outside classroom environments.