TTS (text to speech)

TTS (text to speech) is the technology that turns written words into a spoken voice. In VICIdial, it lets the system say something out loud without you having to record an audio clip first. You type the words, and the dialer reads them to the caller. That sounds small, but it opens up a whole class of messages you could never pre-record.

This is most useful when the words change from call to call. A recorded file can only say one fixed thing, but TTS can read back an order number, a balance, or the caller's name. If you wanted to greet someone by the name from your CNAM (caller ID name) (Caller Name) lookup, TTS is how you'd speak it. Anything that's different for every single caller is a job for text to speech rather than a recording.

Where you'll meet it

• Inside a Call menu, to read out options that you'd rather not re-record every time they change.
• In a Survey dialing flow, to confirm a value the caller just keyed in so they know it landed.
• Through a custom script — often an AGI (Asterisk Gateway Interface) (Asterisk Gateway Interface) program — that pulls live data and speaks it back.

TTS needs an engine running on the box, and the voices vary in quality. For anything a caller hears often — like an Ingroup greeting — a clean recorded file usually still sounds better. Save TTS for the parts that genuinely have to change. Some setups even fetch the text from a URL (in VICIdial) so the spoken message can be updated without touching the dialer.

It matters because it bridges the gap between your live data and your phone calls. A static recording can't say a balance or an appointment time, but TTS can — which means callers get the exact detail they were after without waiting for an agent to look it up.