TTS

TTS, or text-to-speech, is the system that converts a model's text response into spoken audio, giving an avatar a voice that can be streamed and lip-synced in real time.

How TTS works in avatars

TTS stands for Text-to-Speech. It converts the model's text response into spoken audio, giving the avatar a voice that the user can hear in the live session.

For real-time avatars, TTS needs to start quickly and often stream audio as it is generated. Waiting for the full response before speaking can add enough delay to make the avatar feel less conversational.

A concrete example: the model begins answering a support question, and the TTS system starts producing the first spoken words while the rest of the answer is still being generated.

Voice quality matters, but speed and control matter too. A production avatar needs speech that sounds natural, stays aligned to the persona and gives the lip-sync system enough timing information to animate the face.

What Anam ships

Anam's Cara-4 model delivers expressive real-time avatars with around 150 ms server-side avatar-generation latency once a session is running, across 70+ languages. Builders use JavaScript and Python SDKs or integrations for LiveKit, Pipecat, ElevenLabs Agents, Agora, and VideoSDK. Bring any AI stack including OpenAI, Claude, Gemini, Mistral, Groq, Deepgram, Cartesia, or custom providers. The platform supports WebRTC delivery, SOC 2 Type II, HIPAA, zero data retention, and regional data residency. Sessions stream low-latency audio and video to browsers and native apps.

Explore the real-time avatar API →

Related terms

ASR (speech recognition)

ASR, or automatic speech recognition, is the system that converts live spoken audio into text so an avatar or agent can understand the user and decide what to do next.

Lip sync

Lip sync is the process of aligning an avatar's mouth shapes and facial motion with generated speech audio, so the character appears to speak the words at the right time.

AI voice agent

An AI voice agent is a speech-first conversational agent that uses speech recognition, an LLM or agent workflow, and text-to-speech to understand spoken input and reply naturally in real time.

Sub-second latency

Sub-second latency is the end-to-end response time under one second between a user speaking and an avatar replying, covering speech recognition, model reasoning, speech synthesis, animation, and network streaming.

Frequently asked questions

What does TTS do in a real-time avatar session?

TTS turns the model's text response into spoken audio, which the avatar can stream to the user while the face animation and lip sync follow the voice.

Why is streaming TTS important for avatars?

Streaming TTS can begin speaking before the full answer is complete. That lowers perceived latency and keeps the conversation from feeling like a delayed file render.

Is voice quality enough for avatar TTS?

No. Voice quality matters, but so do startup time, pronunciation, prosody, interruption handling, emotional fit, and how reliably the audio drives the face.

How does TTS affect lip sync?

The generated audio gives the avatar timing for mouth shapes and facial motion. If TTS timing or pronunciation is unstable, the lip sync can look delayed or unnatural.

Last updated: 17th July 2026 · Reviewed quarterly.

← Back to glossary