AssemblyAI

Universal model optimized for diverse accents and languages

“og:title”: “Stenox Docs”

This is a TRANSCRIPTION provider. It converts your speech to text. For AI enhancement like grammar correction, see AI Enhancement Providers.

Overview

AssemblyAI provides high-accuracy speech-to-text with their Universal-1 model, specifically optimized for diverse accents, dialects, and multilingual content.

Key Benefits:

• 🌍 Accent-optimized - Excellent for non-native English speakers
• 🎯 High accuracy - Universal-1 model trained on diverse datasets
• 🌐 Multilingual - English, Spanish, French, German, Italian, Portuguese, and more
• 🎁 $50 free credits - No credit card required for signup

Free Tier

New AssemblyAI users get $50 in free credits!

That’s approximately 5,000+ minutes of transcription - several weeks to months of regular use.

No credit card required to sign up and start transcribing.

What You Get

• $50 in credits upon signup
• No credit card required for free tier
• All features included - Universal-1, punctuation, streaming, etc.
• Credits valid for active accounts

After credits are used, pay-as-you-go pricing starts at $0.0009 per minute (very affordable!).

Getting Your Free API Key

1. Visit AssemblyAI

   Go to assemblyai.com and click Get Started Free or Sign Up.

2. Create an account

   Sign up with:

   • Email address and password
   • Or use Google/GitHub sign-in

   No credit card required!

3. Verify your email

   Check your email for a verification link from AssemblyAI and click it to activate your account.

4. Access the Dashboard

   Once logged in, you’ll be redirected to the AssemblyAI Dashboard.

5. Find your API key

   Your API key is displayed prominently on the Dashboard:

   1. Look for the API Key section
   2. Click Copy to copy your key to clipboard
   3. Or click Show to reveal the full key
6. Add key to Stenox

   1. Open Stenox Settings
   2. Go to Transcription tab
   3. Select AssemblyAI as provider
   4. Paste your API key
   5. Choose your model (Universal-1)
   6. Click Save

Keep your API key private! Don’t share it or commit it to public repositories.

Available Models

Universal-1 (Recommended)

AssemblyAI’s flagship model optimized for accuracy across diverse scenarios:

Universal-1 English:

• Trained on diverse English accents (US, UK, Australian, Indian, African, etc.)
• Best for: English dictation with any accent

Universal-1 Multilingual:

• Supports English, Spanish, French, German, Italian, Portuguese
• Auto-language detection
• Best for: Non-English languages or code-switching

Recommendation: Use Universal-1 English for English dictation, or Universal-1 Multilingual if you speak multiple languages.

Language Support

AssemblyAI supports multiple languages with auto-detection:

Supported Languages:

• English - All accents and dialects (US, UK, Australian, Indian, etc.)
• Spanish - Spain and Latin American variants
• French - France and Canadian variants
• German
• Italian
• Portuguese - Portugal and Brazilian variants
• Auto-detect - Automatically identify language being spoken

Configure language in Stenox profile settings.

Configuration in Stenox

1. Open Stenox Settings

   Click Stenox menu bar icon → Settings

2. Navigate to Transcription

   Select the Transcription tab

3. Select AssemblyAI

   Choose AssemblyAI from the provider dropdown

4. Enter your API key

   Paste the API key you copied from AssemblyAI Dashboard

5. Choose model

   Select Universal-1 (recommended)

6. Configure language

   • English only: Select English
   • Other languages: Select Spanish, French, etc.
   • Multiple languages: Select Auto-detect
7. Save

   Click Save to activate AssemblyAI

Features

Accent Optimization

AssemblyAI’s Universal-1 model excels with diverse accents:

• Non-native English speakers
• Regional accents (Southern US, Boston, Scottish, etc.)
• International English (Indian, Nigerian, Singaporean, etc.)
• Mixed accent environments

If you have a strong accent or speak with non-native pronunciation, AssemblyAI often provides better accuracy than other providers.

Streaming Transcription

AssemblyAI uses WebSocket streaming for real-time results:

• Audio streams to AssemblyAI as you speak
• Processing begins immediately
• Results return in < 1-2 seconds after you stop speaking

Automatic Features

Punctuation:

• Automatic periods, commas, question marks
• Natural capitalization
• Sentence structure optimization

Formatting:

• Numbers formatted appropriately (e.g., “twenty-five” → “25”)
• Dates and times formatted correctly
• Currency and units handled intelligently

Performance

Expected processing times with AssemblyAI:

Recording Length	Processing Time
5 seconds	< 1 second
10 seconds	1-2 seconds
30 seconds	2-3 seconds
60 seconds	3-4 seconds

AssemblyAI is optimized for streaming, so longer recordings don’t significantly increase processing time.

Usage Tracking

Monitor your credit usage in the AssemblyAI Dashboard:

1. Visit assemblyai.com and log in
2. Go to Dashboard
3. View:
   • Remaining credits ($50 initially)
   • Usage this month
   • Detailed transcription logs

Stenox does not track AssemblyAI usage. Check your Dashboard for accurate credit balance.

Privacy Considerations

What AssemblyAI sees:

• Your audio is sent to AssemblyAI servers for processing
• Audio and transcripts may be logged for quality improvement
• Data is encrypted in transit (HTTPS/WSS)
• AssemblyAI offers data retention control options

For complete privacy:

• Use WhisperKit (local) for 100% offline processing
• Or review AssemblyAI’s privacy policy and configure data retention settings

When to Use AssemblyAI

Non-native accents

Optimized for diverse English accents and dialects.

Multilingual use

Supports multiple languages with auto-detection.

Reliable accuracy

Consistent quality across different speakers and environments.

Cost-effective cloud

Affordable pricing ($0.0009/min) after free credits.

Troubleshooting

API key not working

• Verify you copied the complete key from AssemblyAI Dashboard
• Check that your account is verified (email confirmation)
• Try generating a new API key in the Dashboard
• Ensure no extra spaces or characters were added

Transcription fails or errors

• Check your internet connection
• Verify you have remaining credits in Dashboard
• Ensure audio is being captured (check Stenox recording overlay)
• Try a shorter test recording first

Poor accuracy for your accent

• AssemblyAI is generally excellent with accents, but try:
  • Speaking slightly more clearly
  • Reducing background noise
  • Using a better microphone
  • Adding custom vocabulary in Stenox
• Compare with DeepGram or WhisperKit for your specific accent

Wrong language detected (auto-detect)

• Instead of auto-detect, manually select your language in Stenox profile settings
• Speak longer phrases (auto-detect needs more context)
• Ensure you’re using Universal-1 Multilingual model

Pricing After Free Credits

Once your $50 in credits are used:

Feature	Price
Standard transcription	$0.0009 per minute
Per hour	~$0.054 per hour

Example monthly costs:

• 1 hour per day = ~$1.62/month
• 30 minutes per day = ~$0.81/month

Very affordable for regular use!

View full pricing →

Comparison: AssemblyAI vs Others

Feature	AssemblyAI	DeepGram	WhisperKit
Provider cost	$50 free credits	$200 free credits	No cost (local)
Speed	Fast (1-2s)	Fastest (< 1s)	Medium (3-5s)
Accent handling	Excellent	Very Good	Good
Multilingual	6 languages	30+ languages	100+ languages
Privacy	Cloud	Cloud	100% local
Ongoing cost	$0.0009/min	$0.0125/min	Free

All providers work within your Stenox plan limits. Lifetime unlocks unlimited daily use with any provider.

Next Steps

Add AI Enhancement

Enhance transcripts with grammar correction and formatting.

Create Profiles

Different providers for different apps or use cases.

Free Tier Guide

Maximize free usage across all providers.

Add Vocabulary

Improve accuracy with custom words and terms.