Skip to main content
Installation RequirementsTo use most channel connectors, you need to install the channels dependency group:
For more information about dependency groups, see our Python Versions and Dependencies reference page.

​
Connecting to a Channel

​
Text & Chat

Learn how to make your assistant available on:

​
Voice

Learn how to build voice assistants with: To learn more about integration ASR and TTS engines, head over to Speech Integrations. For how Rasa spaces consecutive bot audio on voice stream channels (minimum gaps and silence between messages), see Minimum pacing between bot messages (voice streams). Only some channels deliver assistant text incrementally. Voice stream connectors and the REST channel (with ?stream=true) support real-time streaming when paired with a streaming-capable action executor; most messaging channels do not. See Built-in output channels with streaming support.

​
Testing Channels on Your Local Machine

If you’re running a Rasa server on localhost, most external channels won’t be able to find your server URL, since localhost is not open to the internet. To make a port on your local machine publicly available on the internet, you can use ngrok. Alternatively, see this list tracking and comparing other tunneling solutions. After installing ngrok, run:
When you follow the instructions to make your assistant available on a channel, use the ngrok URL. Specifically, wherever the instructions say to use https://<host>:<port>/webhooks/<CHANNEL>/webhook, use <ngrok_url>/webhooks/<CHANNEL>/webhook, replacing <ngrok_url> with the randomly generated URL displayed in your ngrok terminal window. For example, if connecting your bot to Slack, your URL should resemble https://26e7e7744191.ngrok.io/webhooks/slack/webhook.
With the free-tier of ngrok, you can run into limits on how many connections you can make per minute. As of writing this, it is set to 40 connections / minute.
Alternatively you can make your assistant listen on a specific address using the -i command line option:
This is particularly useful when your internet facing machines connect to backend servers using a VPN interface.