AI & Machine Learning API Key

AssemblyAI REST API

Speech-to-Text and Audio Intelligence API

AssemblyAI provides state-of-the-art speech recognition and audio intelligence APIs. Developers use it to transcribe audio and video files, identify speakers, detect topics, analyze sentiment, and extract insights from conversational data. The API powers voice applications, meeting transcription tools, content moderation systems, and media analysis platforms with industry-leading accuracy.

Base URL https://api.assemblyai.com/v2

API Endpoints

Method	Endpoint	Description
POST	`/upload`	Upload a local audio or video file to AssemblyAI's servers for transcription
POST	`/transcript`	Submit a transcription job for an audio or video file URL
GET	`/transcript/{id}`	Retrieve the status and results of a transcription job
GET	`/transcript/{id}/sentences`	Get the transcript broken down into individual sentences with timestamps
GET	`/transcript/{id}/paragraphs`	Get the transcript organized into paragraphs with timestamps
GET	`/transcript/{id}/subtitles/srt`	Export the transcript in SRT subtitle format
GET	`/transcript/{id}/subtitles/vtt`	Export the transcript in WebVTT subtitle format
DELETE	`/transcript/{id}`	Delete a transcript and its associated data from AssemblyAI servers
GET	`/transcript`	List all transcripts associated with your account with pagination
POST	`/lemur/task`	Submit a LeMUR task to apply custom AI prompts to transcribed audio
POST	`/lemur/summary`	Generate an AI summary of transcribed audio content
POST	`/lemur/question-answer`	Ask questions about transcribed audio and get AI-generated answers
POST	`/lemur/action-items`	Extract action items from meeting transcripts using AI
GET	`/lemur/response/{id}`	Retrieve the results of a LeMUR task by request ID
POST	`/realtime/token`	Generate a temporary authentication token for real-time streaming transcription

Code Examples

# Upload and transcribe an audio file
curl -X POST https://api.assemblyai.com/v2/transcript \
  -H 'Authorization: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "audio_url": "https://example.com/audio.mp3",
    "speaker_labels": true,
    "sentiment_analysis": true,
    "auto_chapters": true
  }'

# Check transcription status
curl https://api.assemblyai.com/v2/transcript/TRANSCRIPT_ID \
  -H 'Authorization: YOUR_API_KEY'

// Submit a transcription job
const response = await fetch('https://api.assemblyai.com/v2/transcript', {
  method: 'POST',
  headers: {
    'Authorization': 'YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    audio_url: 'https://example.com/audio.mp3',
    speaker_labels: true,
    sentiment_analysis: true,
    auto_chapters: true
  })
});

const { id } = await response.json();

// Poll for completion
const pollTranscript = async (transcriptId) => {
  const result = await fetch(`https://api.assemblyai.com/v2/transcript/${transcriptId}`, {
    headers: { 'Authorization': 'YOUR_API_KEY' }
  });
  const transcript = await result.json();
  
  if (transcript.status === 'completed') {
    return transcript;
  } else if (transcript.status === 'error') {
    throw new Error(transcript.error);
  }
  
  await new Promise(resolve => setTimeout(resolve, 3000));
  return pollTranscript(transcriptId);
};

const transcript = await pollTranscript(id);
console.log(transcript.text);

import requests
import time

# Submit transcription job
headers = {
    'Authorization': 'YOUR_API_KEY',
    'Content-Type': 'application/json'
}

data = {
    'audio_url': 'https://example.com/audio.mp3',
    'speaker_labels': True,
    'sentiment_analysis': True,
    'auto_chapters': True
}

response = requests.post(
    'https://api.assemblyai.com/v2/transcript',
    headers=headers,
    json=data
)

transcript_id = response.json()['id']

# Poll for completion
while True:
    response = requests.get(
        f'https://api.assemblyai.com/v2/transcript/{transcript_id}',
        headers=headers
    )
    transcript = response.json()
    
    if transcript['status'] == 'completed':
        print(transcript['text'])
        break
    elif transcript['status'] == 'error':
        raise Exception(f"Transcription failed: {transcript['error']}")
    
    time.sleep(3)

Use AssemblyAI from Claude / Cursor / ChatGPT

Get a hosted MCP endpoint for AssemblyAI. Paste your AssemblyAI API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls AssemblyAI directly with your credentials — no local install, works on mobile.

transcribe_audio Transcribe audio or video files from URLs with optional speaker diarization, sentiment analysis, and content moderation

summarize_transcript Generate AI-powered summaries of transcribed audio using LeMUR for meeting notes, podcast summaries, or content briefs

extract_action_items Automatically identify and extract action items, tasks, and follow-ups from meeting transcripts

analyze_audio_sentiment Analyze the emotional tone and sentiment of spoken content across utterances and speakers

generate_subtitles Create SRT or VTT subtitle files from transcribed audio for video accessibility and localization

Connect in 60 seconds

Paste your AssemblyAI key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.

Connect AssemblyAI to your AI →

AssemblyAI REST API

API Endpoints

Sponsor this page

Code Examples

Use AssemblyAI from Claude / Cursor / ChatGPT

Connect in 60 seconds

Related APIs