What is Text-to-Speech (TTS)?
Text-to-speech (TTS) converts written text into spoken audio. In a voice agent it is the speaking step, and its time-to-first-audio and naturalness shape how human the call feels.
Also known as: TTS, speech synthesis, voice synthesis
How it works
A TTS model converts text into a waveform, choosing pronunciation, pitch, pace and pauses. For conversation, streaming synthesis matters most: the system begins playing audio for the first words while the rest of the sentence is still being generated, which cuts the silence the caller hears.
What to check beyond how it sounds
Voice quality is judged in a demo, but three things only show up in production. Pronunciation of numbers, names and currency in the caller's language. Consistency of the same voice over a long call. And behaviour when interrupted: the audio must stop immediately, cleanly, without a trailing syllable.
Common mistake
Choosing the most expressive voice and ignoring script design. A warm voice reading a 60-word paragraph still feels like a recording. Short sentences, one question at a time and natural fillers do more for perceived quality than any model upgrade.
Where this shows up at Vaami
Vaami speaks in 30+ languages, including Hindi, Hinglish, Tamil, Telugu, Kannada, Marathi, Bengali, Gujarati, Punjabi and Malayalam.
See it on a real call.
Hear how a Vaami voice agent handles this on a live conversation in your language and your use case.