File Processor Module
Convert PDFs, images, documents, and media files to AI-ready formats.
Overview
The File Processor is NeuroGen's core document conversion tool. It handles: - PDFs - Extract text, tables, and structure - Images - OCR text extraction - Documents - Word, Excel, PowerPoint conversion - Media - Audio/video transcription
Getting Started
- Navigate to Dashboard > File Processor
- Upload files by drag-and-drop or file browser
- Configure processing options
- Click Process
- View results and download outputs
Supported File Types
| Category | Formats |
|---|---|
| Documents | PDF, DOCX, DOC, ODT, RTF, TXT |
| Spreadsheets | XLSX, XLS, CSV |
| Presentations | PPTX, PPT, ODP |
| Images | PNG, JPG, JPEG, TIFF, BMP, GIF |
| Audio | MP3, WAV, M4A, FLAC, OGG |
| Video | MP4, MOV, AVI, MKV, WEBM |
Processing Options
PDF Processing
- Extract text - Full text extraction with formatting
- Extract tables - Convert tables to structured data
- OCR mode - For scanned/image PDFs
- Page range - Process specific pages only
Image OCR
- Language - Select OCR language(s)
- Enhancement - Auto-enhance before OCR
- Output format - Plain text or structured
Audio/Video
- Transcription - Speech-to-text conversion
- Language - Specify audio language
- Speaker diarization - Identify speakers (premium)
Output Formats
| Format | Use Case |
|---|---|
| JSON | Structured data with metadata |
| JSONL | Training data for LLMs |
| Markdown | Readable documents |
| CSV | Spreadsheet data |
| Plain Text | Simple text extraction |
Session Output
Each processing job creates a session containing:
- output.json - Full structured output
- text_content.txt - Extracted text
- metadata.json - Processing metadata
- Additional format-specific files
Tips for Best Results
- Clean PDFs work best - Native PDFs process faster than scanned
- Use OCR for scans - Enable OCR mode for image-based PDFs
- Check language settings - Set correct language for OCR
- Preview before share - Verify output quality
Example: Convert PDF for Custom GPT
- Upload your PDF document
- Select JSONL output format
- Process the file
- Go to My Files > Sessions
- Click Share on the output file
- Select Full Access permission
- Use the share link in your Custom GPT
Troubleshooting
"Processing failed"
- Check file isn't corrupted
- Verify file size is within limits (50MB)
- Try a different output format
"No text extracted"
- Enable OCR mode for scanned documents
- Check the document contains readable text
- Try with image enhancement enabled
"Slow processing"
- Large files take longer
- OCR is slower than native text extraction
- Audio/video transcription can take several minutes
Files automatically appear in My Files > Sessions after processing