You never get a quiet moment at an airport or a railway station, from the chatter to the sound of rolling suitcases.
On top come all the announcements about security, connections, delays and missing persons.
Whatever the topic, recorded announcements are usually made in women's voices, which are universally preferred, surveys show. Also, the messages are conveyed in calm and friendly tones, rather than the somewhat harsher commands you sometimes hear in live announcements.
Nonetheless, often it seems passengers would prefer quiet and at Frankfurt Airport, the largest and busiest in Germany, operator Frankfurt Fraport pursues a "silent airport" policy.
That means keeping a strict limit on boarding calls and ensuring announcements are only heard within a specific area, even though they are all coordinated and controlled centrally.
Travellers occasionally do still hear announcements made by human beings, but these days it is far more common to hear machine-generated announcements, such as during the boarding process.
The advantage for the airlines is that machines can relieve the burden on staff and significantly speed up the boarding process.
Lufthansa is one of many carriers that prefers this approach. "It means that clear, high-quality announcements are available in different languages and our customers receive standardized, structured information," says Germany's flagship airline.
"We use a system that has stored text modules and various announcement contents," a Fraport spokesman told dpa.
Using standardized announcements helps with planning when and where the announcements should be heard, he said.
Sittig Technologies, a company based near Frankfurt Airport, has interconnected more than 10 autonomous public address systems at the airport and handles the control system.
The system triggers more than 2,500 announcements in more than 24 languages every day.
Ensuring those statements reach passengers' ears involves feeding the voice into the infrastructure at the airport.
"In the vast majority of airports, the public address systems have grown historically," says Thomas Sittig, one of the two managing directors.
The family-run company long since gave up on the idea of developing its own loudspeaker systems, which would put it in competition with industrial giants like Bosch or Honeywell.
Instead, the smaller company uses whatever audio infrastructure is available at an airport, no matter what the manufacturer, and manages the system using standardized software.
The airports they work with are primarily in German-speaking countries, with customers in Berlin, Zurich, Munich and Stuttgart alongside Frankfurt.
They also provide automatic announcements for train operators and clients include Swiss Südostbahn and Sydney Trains, with 300 railway stations.
At airports, it is primarily the airlines who are driving the development of the technology.
"Our business is not the content, but the technology," says sales chief Johannes Sittig. Aided by a grant from a state digital fund, the company used the time during the pandemic to take its PAXGuide technology to a new level, namely to the global data cloud.
Rather than using costly local servers, customers now only need to install a small interface box to retrieve automated announcements securely from the cloud in a matter of seconds.
With around 40 employees, the Sittigs are now targeting the US market and are eyeing around 100 airports on the North American continent, most of which lack the manpower to continue making individual announcements.
They have launched projects with Lufthansa in Los Angeles and with budget airline Frontier in Cincinnati.
Unlike real people, the voices from the cloud stick to scripts and length specifications when making their announcements. "The aim of good announcements is to reduce stress for everyone involved," says co-boss Christian Tischler.
But despite the involvement of technology, behind the scenes, IT staff are constantly tweaking and changing the announcements.
Sittig's team of speakers can generate announcements in more than 40 languages. For each language, trained spokespeople record up to 5,000 text modules in a studio in Cologne.
Some of the speakers may be familiar to travellers, such as actress Alison Rippier whose voice can be heard in Frankfurt.
But there are also new text-to-speech machines from Amazon or Google, for example, which Sittig also offers depending on customer requirements.
However, whether the noise is made by man or machine, the general feeling among human beings still may be that less is more.
