fast elbow
I'm trying to add dictate functionality to a notes application written in C#, MVC and HTML views & javascript (no python, no node).
Got it working using MediaRecorder and WebSocket but the resultant text is a long unformatted stream.
Are there javascript WebSocket parameters I can use to implement smart-format?
What are the available WebSocket parameters?
shell trellisBOT
Hi @fast elbow. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Yes, you can use the smart_format parameter with the WebSocket API to implement smart formatting in your application. This parameter can be included in the WebSocket URL when establishing a connection.
Here is an example of how to use it in JavaScript:
const socket = new WebSocket('wss://api.deepgram.com/v1/listen?smart_format=true&model=nova-2-ea', ['token', 'xxxxxxx']);
In the above example, smart_format=true is used to enable smart formatting. The model parameter is also used to specify the model to be used for transcription.
Please note that the smart_format parameter does punctuation and paragraphs for you, so you shouldn't need to use both the smart_format and paragraph and punctuation options in this request source (https://github.com/orgs/deepgram/discussions/635).