Documentation – LRC Generator & Lyrics File Maker
AI Lyrics Recognition
AI Lyrics Recognition
Automatically recognize lyrics from audio/video files using AI and generate precisely synchronized LRC files.
The core feature of AI LRC Generator is using advanced speech recognition and natural language processing to automatically transcribe vocals from audio/video into timestamped LRC files.
Features
Multi-language Support
- English, Chinese (Simplified/Traditional), Korean, Japanese
- Intelligent language detection, automatically matching the optimal recognition model
- Supports mixed-language content (e.g., Korean-English mixed K-Pop songs)
High-precision Time Sync
- Millisecond-level timestamp alignment
- Automatic beat matching between lyrics and audio/video
- Speech recognition in complex background music environments
Batch Processing
- Premium users can upload multiple audio/video files at once
- Process an entire album or playlist in one go
- Batch download all generated LRC files
Usage Steps
Prepare Audio/Video Files
For optimal recognition results, we recommend:
- High-quality audio/video: 320kbps or lossless format
- Clear vocals: Minimize background noise as much as possible
- Complete songs: Avoid clipped or corrupted files
Upload and Analyze
- Go to the AI LRC Generator page
- Upload an audio/video file; the system automatically detects duration
- Confirm credit consumption (2 credits per minute of media)
- Click "Start Recognition"
Wait for Processing
The system goes through the following statuses:
| Status | Description |
|---|---|
PENDING | File uploaded, waiting for analysis trigger |
QUEUED | Entered processing queue, shows current queue position |
PROCESSING | AI is analyzing audio/video content |
COMPLETED | Processing complete, results can be viewed and downloaded |
FAILED | Processing failed, retry after checking the error reason |
View and Edit Results
After recognition is complete, you can:
- Preview the sync effect between lyrics and audio/video online
- Enter the editor to correct recognition errors
- Adjust timeline offsets
- Export to multiple formats
Credit Calculation Rules
Credits are calculated based on media duration, with any duration under 1 minute rounded up to 1 minute:
Credits needed = ceil(duration in seconds / 60) × 2
Examples:
- 3 min 20 sec = 4 minutes × 2 = 8 credits
- 1 min 05 sec = 2 minutes × 2 = 4 credits
Best Practices
Improve Recognition Accuracy
- Select the correct language setting
- Use studio-quality audio/video
- Avoid overly compressed or distorted files
- For multi-person choruses, try the lead vocal track first
Troubleshooting Failures
| Problem | Solution |
|---|---|
| File too large | Compress audio/video or upgrade plan |
| Insufficient credits | Go to Pricing to recharge or upgrade |
| Large recognition deviation | Check if language setting is correct |
| Instrumental music without lyrics | This tool only supports media containing vocals |
Processed audio/video files are retained on the server for a period for your download and preview, after which they are automatically cleaned to ensure privacy and security.