Sonartext STT Actor - Professional Speech-to-Text Transcription

Transform audio content into high-quality text transcriptions with enterprise-grade accuracy using state-of-the-art AI models. Perfect for content creators, researchers, businesses, and developers who need reliable speech-to-text conversion.

🚀 Key Features

• 🎯 State-of-the-Art AI: Powered by Whisper-v3-large-turbo for industry-leading accuracy
• 👥 Speaker Diarization: Automatically identify and separate different speakers
• 📁 Multiple Input Sources: Support for direct uploads, YouTube videos, Twitter audio, Google Drive, AWS S3, and URLs
• ⚡ Fast Processing: Optimized for speed without compromising quality
• 💰 Cost Control: Built-in cost estimation and limits to manage expenses
• 📊 Flexible Output: Multiple formats including JSON, plain text, SRT, and VTT
• 🔐 Enterprise Security: Secure API integration with proper error handling
• 📈 Usage Analytics: Detailed processing statistics and performance metrics

🎯 Use Cases

Content Creation & Media

• Podcast Transcription: Convert episodes to searchable text for show notes and SEO
• Video Subtitles: Generate accurate captions for YouTube, social media, and streaming
• Interview Processing: Transcribe interviews with automatic speaker identification
• Course Materials: Create accessible text versions of educational audio content

Business & Research

• Meeting Transcription: Document important discussions with speaker attribution
• Research Analysis: Convert audio interviews and focus groups to analyzable text
• Call Center Analytics: Process customer service calls for quality assurance
• Legal Documentation: Transcribe depositions and hearings with high accuracy

Accessibility & Compliance

• ADA Compliance: Provide text alternatives for audio content
• Multi-language Support: Transcribe content in multiple languages
• Hearing Accessibility: Make audio content accessible to deaf and hard-of-hearing users

📥 Input Configuration

Simple Input Fields

No complex JSON required! Just fill out the form fields:

• Input Method: Choose how to provide your audio (Upload File, URL, YouTube, etc.)
• Audio File/URL: Provide the file or URL based on your chosen method
• Language: Select language or leave blank for auto-detection
• Timestamps: Choose timestamp detail level
• Speaker Diarization: Enable to identify different speakers
• Response Format: JSON, Plain Text, SRT, or VTT subtitles

Example Inputs

File Upload

Input Method: Upload File
Audio File: [Choose your file...]
Language: English
Response Format: JSON

YouTube Video

Input Method: YouTube Video
YouTube URL: https://youtube.com/watch?v=abc123
Language: Auto-detect
Timestamps: Segment-level
Response Format: SRT Subtitles

URL Processing

Input Method: Direct URL
File URL: https://example.com/audio.mp3
Speaker Diarization: Yes
Min Speakers: 2
Max Speakers: 4

Input Source Options

Input Method	Description	What You Provide
Upload File	Upload audio/video from your computer	Click "Choose file" and select your audio/video file
Direct URL	Audio/video file from any web URL	Enter the direct URL to the file
YouTube Video	Extract audio from YouTube video	YouTube video URL (e.g. youtube.com/watch?v=...)
Twitter/X Video	Extract audio from Twitter/X post	Twitter post URL with video
Google Drive	File shared via Google Drive	Public Google Drive file link
AWS S3	File stored in S3 bucket	S3 URL or presigned URL

Configuration Options

Setting	Options	Description
Language	Auto-detect, English, Spanish, French, etc.	Language of the audio content
Timestamps	None, Segment-level, Word-level, Both	Detail level of timestamps
Speaker Diarization	On/Off	Identify different speakers
Min/Max Speakers	1-20	Expected number of speakers (if diarization enabled)
Response Format	JSON, Plain Text, SRT, VTT	Output format for results
Cost Limit	1-10000 cents	Optional spending limit for safety

📤 Output Structure

JSON Format (Default)

{
  "success": true,
  "result": {
    "transcription": "Hello, welcome to our podcast. Today we're discussing AI innovations.",
    "speakers": [
      {
        "speaker": "Speaker 1",
        "start": 0.0,
        "end": 3.2,
        "text": "Hello, welcome to our podcast."
      },
      {
        "speaker": "Speaker 2", 
        "start": 3.5,
        "end": 7.8,
        "text": "Today we're discussing AI innovations."
      }
    ],
    "metadata": {
      "duration": 45.6,
      "wordCount": 124,
      "confidence": 0.94,
      "language": "en"
    }
  },
  "usage": {
    "costCents": 23,
    "processingTime": 12.5,
    "audioMinutes": 0.76
  }
}

SRT Format Example

1
00:00:00,000 --> 00:00:03,200
Hello, welcome to our podcast.

2
00:00:03,500 --> 00:00:07,800
Today we're discussing AI innovations.

Error Handling

{
  "success": false,
  "error": {
    "code": "AUDIO_TOO_LONG",
    "message": "Audio duration exceeds maximum limit",
    "details": {
      "maxDuration": 3600,
      "actualDuration": 4200
    }
  }
}

🚀 Getting Started

1. Basic Usage

1. Choose "Direct URL" as Input Method
2. Enter: https://example.com/meeting-recording.mp3
3. Enable Speaker Diarization: Yes
4. Click "Run"

2. File Upload Usage

1. Choose "Upload File" as Input Method  
2. Click "Choose file" and select your audio
3. Set Language to "Auto-detect"
4. Choose Response Format: "JSON"
5. Click "Run"

3. Run the Actor

Execute the actor and retrieve your transcription from the dataset.

💡 Pro Tips

Optimize Audio Quality

• Clear Audio: Use high-quality recordings with minimal background noise
• Proper Levels: Ensure consistent audio levels across speakers
• Format: MP3, WAV, and FLAC formats work best

Speaker Diarization Best Practices

• Enable for multi-speaker content (meetings, interviews, podcasts)
• Works best with 2-10 distinct speakers
• Clearer speaker separation improves accuracy

Cost Management

• Set maxCostCents to control expenses
• Use cost estimation feature before processing large files
• Consider audio length: pricing is typically per minute

YouTube Processing

• Supports both public and unlisted videos
• Automatically extracts best quality audio
• Handles various video lengths efficiently

💰 Pricing

This actor uses Sonartext's professional STT service with transparent, usage-based pricing:

• Event Based Pricing: $0.01 base fee per Actor Start
• Usage Fees: Pay $0.004 per minute of audio processed
• Volume Discounts: Available for high-usage customers
• Cost Control: Set maximum spending limits per job

🔒 Security & Privacy

• API Security: All requests use secure HTTPS with API key authentication
• Data Privacy: Audio files are processed securely and not stored permanently
• Compliance: Built for enterprise use with privacy-first architecture
• Rate Limiting: Respectful API usage with built-in rate limiting

⚠️ Legal Considerations

• Content Rights: Ensure you have proper rights to transcribe the audio content
• Privacy Compliance: Follow applicable privacy laws (GDPR, CCPA, etc.) when processing personal data
• Platform Terms: Respect YouTube's Terms of Service and other platform policies
• Copyright: Be mindful of copyrighted content when processing media files

🛠️ Advanced Features

Batch Processing

Process multiple files efficiently by chaining multiple actor runs or using the batch processing capabilities.

Integration Options

• APIs: Direct REST API integration available
• Webhooks: Real-time processing notifications
• Zapier: No-code workflow integration
• Custom Solutions: Enterprise integrations available

Supported Audio Formats

• Audio: MP3, WAV, FLAC, AAC, OGG, M4A
• Video: MP4, AVI, MOV, MKV (audio extracted)
• Containers: Support for most common audio/video containers

📞 Support & Documentation

Help & Support

• Documentation: Full API Documentation
• Support: Contact support through the Apify platform
• Community: Join our Discord for community support
• Enterprise: Dedicated support for business customers

Troubleshooting

• Rate Limits: Wait and retry if you hit API limits
• Large Files: Consider breaking very large files into segments
• Quality Issues: Check audio quality and format compatibility
• Costs: Monitor usage through the actor's cost reporting features

Updates & Changelog

This actor is actively maintained with regular updates for:

• New AI model versions
• Additional output formats
• Performance improvements
• Enhanced error handling

---

📊 Actor Metrics

• Response Time: < 30 seconds for typical audio files
• Accuracy: 95%+ for clear audio with proper formatting
• Supported Languages: 50+ languages supported
• File Size Limit: Up to 500MB per file
• Duration Limit: Up to 6 hours per audio file

---

Developed by: Sonartext
Version: 1.0.0
Last Updated: September 2025
License: MIT

For technical issues or feature requests, please use the Apify platform's support system or visit our documentation.