Skip to main content
The pipeline isn’t limited to voice. You can feed it text and have it reply with text or speech, which is useful for chat widgets, omnichannel agents that share one brain across voice and chat, and for testing an agent from the terminal.

Chat

Send text in and receive text or voice out.

Transformation

Rewrite the text stream before TTS: pronunciation, filtering, normalization.

​
Text-capable pipelines

Drop the STT and/or TTS stages to run text in or out: See Pipeline Modes for how the pipeline auto-detects the configuration from the components you pass.

​
References

​
SDK Reference

ChatContext

ChatContext in the Python API reference.

Agent

Agent in the Python API reference.