OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Deepgram Js SDK

Deepgram Js SDK is an open-source JavaScript and TypeScript software development kit for connecting applications to Deepgram automated speech-recognition, text-to-speech and language-understanding APIs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deepgram Js SDK — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when application code needs a JavaScript/TypeScript client for Deepgram APIs, particularly for streaming speech workflows.

Who it is for

  • JavaScript and TypeScript developers adding speech or language API capabilities to applications.
  • Teams implementing live audio transcription or text-to-speech streaming.

Documented uses

  • Send live audio to a speech-recognition WebSocket and receive result messages.
  • Integrate transcription for prerecorded audio files.
  • Connect application workflows to text-to-speech and language-understanding APIs.
Before you choose

Requirements & limits

The repository is public and identifies the SDK as MIT-licensed. The README specifies Node.js 18+ and installation through npm.

  • The SDK covers speech recognition, text-to-speech and language-understanding API use.
  • Streaming connections start closed; register handlers, connect, and wait for the connection to open before sending data.
  • Browser WebSocket use is documented, while browser REST calls require a proxy because of CORS restrictions.
  • Async iteration supports one active iterator per connection and has a queue cap of 1,000 messages or 16 MiB.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a live transcription WebSocket connection

Connect an application to the documented Listen v1 streaming speech-recognition client and send audio.

Before you start

  • The installed @deepgram/sdk package
  • A Deepgram API key configured for the client
  • Audio data to send

Steps

  1. Create a DeepgramClient and call client.listen.v1.connect() with options such as a model and language.

  2. Register connection event handlers, such as a message handler for Results.

  3. Call connect() and await waitForOpen().

  4. Send audio data with sendMedia().

Expected outcome

The application is set up to send live audio and receive speech-recognition result messages through the WebSocket connection.

Guide limitations

  • The guide documents the client sequence, not a verified runtime result.
  • The README states that the connection starts closed and must be connected before sending data.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
deepgram/deepgram-js-sdk · Official homepage title · retained in its source language
How we review sourcesSuggest a correction