Intelligibility and Space-based Voice with Relaxed Delay Constraints

In this paper, we leverage the long range end-to-end scenarios envisioned for lunar and beyond voice conversations by allowing non-traditional additional processing delay for quasi-real-time voice conversations. The concept of improving the quality of end-to-end voice conversations for long delay environments is considered by utilizing Luby transforms on the conjugate-structure algebraic-code-excited linear-prediction (CS-ACELP) codec. In addition, this paper also examines the use of automated speech recognition software as a means of generating a quantifiable metric for speech intelligibility in the spirit of the diagnostic rhyme test (DRT).