assemblyai

---

AssemblyAI Node.js SDK

The AssemblyAI Node.js SDK provides an easy-to-use interface for interacting with the AssemblyAI API, which supports async and real-time transcription, as well as the latest LeMUR models.

Installation

You can install the AssemblyAI SDK by running:

npm install assemblyai

yarn add assemblyai

pnpm add assemblyai

bun add assemblyai

Usage

Import the AssemblyAI package and create an AssemblyAI object with your API key:

import AssemblyAI from "assemblyai";

const client = new AssemblyAI({
  apiKey: process.env.ASSEMBLYAI_API_KEY,
})

You can now use the client object to interact with the AssemblyAI API.

Create a transcript

When you create a transcript, you can either pass in a URL to an audio file, or upload a file directly.

// Using a remote URL
const transcript = await client.transcripts.create({
  audio_url: 'https://storage.googleapis.com/aai-web-samples/espn-bears.m4a',
})

// Uploading a file
const transcript = await client.transcripts.create({
  audio_url: './news.mp4',
})

By default, when you create a transcript, it'll be polled until the status is completed or error. You can configure whether to poll, the polling interval, and polling timeout using these options:

const transcript = await client.transcripts.create({
  audio_url: 'https://storage.googleapis.com/aai-web-samples/espn-bears.m4a',
},
{
  // Enable or disable polling. Defaults to true.
  poll: true,
  // How frequently the transcript is polled in ms. Defaults to 3000.
  pollingInterval: 1000,
  // How long to wait in ms until the "Polling timeout" error is thrown. Defaults to 180000.
  pollingTimeout: 5000,
})

Get a transcript

This will return the transcript object in its current state. If the transcript is still processing, the status field will be queued or processing. Once the transcript is complete, the status field will be completed.

const transcript = await client.transcripts.get(transcript.id)

List transcripts

This will return a paged list of transcripts that you have transcript.

const page = await client.transcripts.list()

You can also paginate over all pages.

let nextPageUrl: string | null = null;
do {
  const page = await client.transcripts.list(nextPageUrl)
  nextPageUrl = page.page_details.next_url
} while(nextPageUrl !== null)

Delete a transcript

const res = await client.transcripts.delete(transcript.id)

Use LeMUR

Call LeMUR endpoints to summarize, ask questions, generate action items, or run a custom task.

Custom Summary:

const { response } = await client.lemur.summary({
  transcript_ids: ['0d295578-8c75-421a-885a-2c487f188927'],
  answer_format: 'one sentence',
  context: {
    speakers: ['Alex', 'Bob'],
  }
})

Question & Answer:

const { response } = await client.lemur.questionAnswer({
  transcript_ids: ['0d295578-8c75-421a-885a-2c487f188927'],
  questions: [
    {
      question: 'What are they discussing?',
      answer_format: 'text',
    }
  ]
})

Action Items:

const { response } = await client.lemur.actionItems({
  transcript_ids: ['0d295578-8c75-421a-885a-2c487f188927'],
})

Custom Task:

const { response } = await client.lemur.task({
  transcript_ids: ['0d295578-8c75-421a-885a-2c487f188927'],
  prompt: 'Write a haiku about this conversation.',
})

Transcribe in real time

Create the real-time service.

const service = client.realtime.createService();

You can also pass in the following options.

const service = client.realtime.createService({
  realtimeUrl: 'wss://localhost/override',
  apiKey: process.env.ASSEMBLYAI_API_KEY // The API key passed to `AssemblyAI` will be used by default,
  sampleRate: 16_000,
  wordBoost: ['foo', 'bar']
});

You can also generate a temporary auth token for real-time.

const token = await client.realtime.createTemporaryToken({ expires_in = 60 });
const rt = client.realtime.createService({
  token: token
});

[!WARNING] Storing your API key in client-facing applications exposes your API key. Generate a temporary auth token on the server and pass it to your client.

You can configure the following events.

rt.on("open", ({ sessionId, expiresAt }) => console.log('Session ID:', sessionId, 'Expires at:', expiresAt));
rt.on("close", (code: number, reason: string) => console.log('Closed', code, reason));
rt.on("transcript", (transcript: TranscriptMessage) => console.log('Transcript:', transcript));
rt.on("transcript.partial", (transcript: PartialTranscriptMessage) => console.log('Partial transcript:', transcript));
rt.on("transcript.final", (transcript: FinalTranscriptMessage) => console.log('Final transcript:', transcript));
rt.on("error", (error: Error) => console.error('Error', error));

After configuring your events, connect to the server.

await rt.connect();

Send audio data.

// Pseudo code for getting audio
getAudio((chunk) => {
  rt.sendAudio(chunk);
});

Close the connection when you're finished.

await rt.close();

Tests

To run the test suite, first install the dependencies, then run pnpm test:

pnpm install
pnpm test