AI Glossary / TTS / STT
🎨 Image, video and audio
TTS / STT
Text-to-speech (text→voice) and speech-to-text (speech→text). Narration and transcription.
TTS (text-to-speech) converts text to speech - it's the voice of narrators, audiobooks, and assistants. Modern TTS sounds natural, with intonation and emotion, also in English, with proper pronunciation and accent. STT (speech-to-text), also called ASR, works the other way: converts speech recordings to text, powering meeting transcriptions, captions, and dictation. Together they create a complete voice loop: STT listens, the language model thinks, TTS responds - that's how voice conversations with chatbots work. Quality in English varies between tools, so it's worth testing on your own recordings.
Related concepts