Early reflections within 30-40 ms improve speech intelligibility
Evidence
In the field of architectural acoustics it has long been recognized that early reflections improve speech intelligibility. For this to happen, they must arrive within an “integration interval” within which there is an effective amplification of the direct sound—it is perceived to be louder. For speech, single reflections at the same level as the direct sound contribute usefully to the effective sound level, and thereby the intelligibility, up to about a 30 ms delay. For delayed sounds 5 dB lower than the direct sound, the integration interval is about 40 ms. These intervals embrace all consequential early reflections in domestic rooms or control rooms, although in larger rooms longer delays can be problematic (Lochner and Burger, 1958). — , Toole, Floyd E - Sound reproduction-…, 824
Germane to this discussion are the findings that for speech intelligibility—a crucial consideration for movies—it is the early reflections that are the main contributors. Early reflection energy arriving within about the first 50 ms following the direct sound has the same effect on speech intelligibility scores as an equal increase in the direct sound energy. This was true for both normal and hearing-impaired listeners. — Toole, Floyd E, Sound reproduction- the acoustics and…, 1251
Categorization
- XREF: Directly reverses the common audio-community belief (flagged at 823) that reflected sound hurts dialog clarity; ties to the Haas/precedence effect literature.” — connects to: Ground reflection colors all outdoor speech and creates comb-filter pitch; Early room reflections benefit speech intelligibility as much as raised direct sound level — Toole, Floyd E, Sound reproduction- the acoustics and…, 557; The 30 ms fusion limit applies only to equal-level speech in hemi-anechoic space (+1 more)
- CANDY: A crisp falsifiable number — reflections up to 40ms make speech clearer, not muddier — great conversation opener.
- QUESTION: How does this early reflection benefit differ between speech and music, or between listeners using binaural hearing aids?
- corroborated: 2 independent passages support this claim — , Toole, Floyd E - Sound reproduction-…, 824; Toole, Floyd E, Sound reproduction- the acoustics and…, 1251