Skip to content
ReviewRamble

The weekly brief

AI, reviewed, in your inbox.

  • New reviews of tools, models, MCP servers and repos, with the verdict up front
  • Re-checked prices and ratings when something changes
  • New analysis pieces as they are published

Models, agents, MCP servers and the tools actually worth your time.

One email a week. Unsubscribe with the link in any brief. How we use your email

Text-to-speech

Text-to-speech (TTS) is software that turns written text into spoken audio. Modern TTS uses neural networks that predict the sound of speech from the text, then render it as a waveform.

Also called: TTS, speech synthesis, AI voice generator

By Divya Sharma · Checked Nov 15, 2026 · How we review

TTS is the voice in screen readers, navigation apps, audiobooks and voice agents. Older systems glued together recorded fragments of speech. Neural TTS instead learns from hours of recorded voice how text should sound.

How it works

Google's Tacotron 2 shows the classic two-stage design. One network maps characters to a mel spectrogram, a picture of how sound energy changes over time. A second network, the vocoder, turns that picture into audio. Many newer models skip the spectrogram and predict audio tokens directly, the same way a language model predicts words.

Choosing a TTS model

The things that matter to builders are naturalness, latency to the first audio, language coverage, control over pace and emotion, and price per character. Hosted services such as ElevenLabs compete with small open models such as Kokoro 82M that run on a laptop.

Related terms

TTS is the reverse of speech-to-text. Making TTS sound like one specific person is voice cloning. A speech-to-speech model can replace the text step entirely in a live conversation.

Limits

TTS can mispronounce names, numbers and abbreviations, and long passages can drift in tone. Output that sounds human can be misused to impersonate people.

Related terms

Speech-to-text
Speech-to-text, also called automatic speech recognition (ASR), is software that converts spoken audio into written text. It powers transcription, captions, voice commands and the listening half of voice agents.
Voice cloning
Voice cloning is the use of AI to make a text-to-speech system speak in a specific person's voice, learned from a recording of that person. Some systems need only a short sample.
Speech-to-speech model
A speech-to-speech model takes spoken audio in and produces spoken audio out directly, without first converting the speech to text and back. It is used for live voice conversation and translation.
Voice agent
A voice agent is an AI agent that holds a spoken conversation in real time, listening to a person, deciding what to say or do, and answering in a synthetic voice, often over a phone line or in an app.
Text-to-music
Text-to-music is the task of generating music audio from a written description, such as genre, mood, instruments and tempo. Some systems also write and sing lyrics.

Sources