My question is how homilies were given in gigantic basilicas and cathedrals like San Pietro before microphones were invented… the acoustics can make a pronounced speaking voice loud enough, but not very clear.
I am a professional acoustician.
I can tell you very simply: If we are designing a room to have good speech inteligibity we aim to have a very low reverberation time. A average, furnished, carpeted living room might be about 0.5 seconds.
A classroom for primary school children in the UK is required to be 0.4 seconds or less. - particularly if there are to be hearing impaired or other special needs children in the room.
otherwise it’s 0.6 seconds. Secondary school (High School) may have up to 0.8 seconds, and dedicated music classrooms may have around 1 second.
Video Conferencing suites (where there’s a video link in a boardroom) are designed to be almost anechoic (almost no reverb at all). (0.1 - 0.2 seconds)
A Basilica, traditional large chapel or cathedral is likely to have a very long reverberation time of between 5 - 7 seconds. some may even top 12 seconds.
A “Reverberation time” is the time between a sound stopping at source, and it’s effects in the room to be 60 dB below the original level.
Now in order to understand speech, it is essential that the sound of each consonant is audible above the background noise which includes the echoes of the previous vowel and consonant sounds. if we think of the word “Back” there are 2 transient sounds: “Ba” and “ck”. the plosive ck is very much quieter that the vowel sound in “Ba”. The difference in sound level is about 25 dB vor normal voice and more for a raised voice. (The ck sound is unvoiced and almost unaffected by speaking louder)
The time between the end of the “Ba” and the very short “ck” in normal speech is about .35 of a second.
If the Rev. Time of a room is about .5 of a second, and the listener is not very close to the speaker, the sound of the “Ba” will have dropped by those 25 dB before the “ck” sound is made. In a Rev. Time of 1.5 seconds, it will still be about 10 dB louder, rendering the “ck” inaudible.
In very long reverb times entire short words get lost, and the differences between vowel sounds become muddled in the mush of sound.
Large Cathedrals and Basilicas were designed for Gregorian Chant. This deliberately slow form of chanting takes advantage of the build-up of noise in the echoy environment and aids understanding of the words therin.
similarly this is also what gives rise to the traditional image of the vicar, or priest with a Very slow, and overly enunciated speech pattern, as mocked by movies like the marriage scene in “The Princess Bride” and other similar parodies.
This slow speech is deliberately taught to imporve intelegibility when speaking to a group in a reverberant room. it allows more time for the sound to decay between each phoneme.
Traditional Church Design, included a pulpit which was raised up, near or among the congregation. It has a solid surface directly behind the priests position. This creates an echo from so close to the speaker that it is indistinguishable from the original sound (to the human ear) this boost the sound of the priests voice by 3dB.
Ditto, in the Ad Orientum altar, the solid wall contianing the tabernacle reflects and amplifies the priests voice.
These benefits are lost with the modern altar, and the modern lectern
Modern priests rarely use Chant.
Our Choirs ditto rarely use chant.
_