Skip to main content

AssemblyAI plugin for Rivet

Rivet is an open-source visual AI programming environment. Through a collaboration between AssemblyAI and Rivet, you can use AssemblyAI speech-to-text and LeMUR capabilities in Rivet.

Quickstart

  1. 1

    In your Rivet project, switch to the Plugins tab, find the AssemblyAI plugin and click Add.

    AssemblyAI plugin in the Rivet Plugins tab.

  2. 2

    Next, swap back to the Canvas tab, click the three-dotted button, and then click Settings in the context menu. In the Settings dialog, switch to the Plugins tab, and enter your AssemblyAI API key in the AssemblyAI API Key field.

    AssemblyAI API key field in Rivet Settings.

  3. 3

    Now you can add AssemblyAI nodes to your canvas by right-clicking on the canvas, and select any of the nodes under the AssemblyAI category.

note

At the moment you must use the node executor to use the AssemblyAI plugin in the Rivet UI. You may also use the plugin when embedding Rivet in your own application.

Nodes

Transcribe Audio node

The Transcribe Audio node transcribes audio using the AssemblyAI API. It will return a transcript of the given audio source.

Transcribe Audio node

LeMUR nodes

LeMUR is a framework by AssemblyAI to process audio files with an LLM. The AssemblyAI plugin has a dedicated node for each LeMUR endpoint. Each node accepts Transcript IDs as input which you can get from the Transcribe Audio node. Additional parameters are available as inputs and as node configuration. For more information what these parameters do, see LeMUR API reference.

LeMUR Summary node

The LeMUR Summary node uses LeMUR to summarize a given transcript.

LeMUR Summary node

LeMUR Q&A node

LeMUR can generate answers from a transcript and questions.

LeMUR Question & Answer node

LeMUR Custom Task node

LeMUR can generate a response from a prompt and transcript.

LeMUR Custom Task node

LeMUR Action Items node

The LeMUR Action Items node returns a list of action items from a meeting transcript.

LeMUR Action Items node

Additional resources

You can learn more about using Rivet with AssemblyAI in these resources: