Quick Start
Import your first video, run an analysis, and generate a preliminary report in under 10 minutes.
Quick Start Guide#
This guide walks you through the core workflow of FrameCounsel: importing video evidence, running automated analysis, and reviewing findings. By the end, you will have a working case project with transcribed footage and an initial contradiction report.
Create a New Case#
Open FrameCounsel and select File > New Case or press . Enter the following details:
- Case Name - A descriptive identifier (e.g., "State v. Martinez - BWC Footage")
- Case Number - Your firm's internal reference number
- Client Name - The defendant's name (encrypted at rest)
- Jurisdiction - Relevant court jurisdiction for report formatting
Case Encryption
All case data is encrypted using AES-256 at rest. Case metadata, transcripts, and analysis results are stored in an encrypted SQLite database within your workspace directory.
Import Video Evidence#
With your case open, click the Import button in the toolbar or drag video files directly into the evidence panel. FrameCounsel supports the following formats:
- Video: MP4, MOV, AVI, MKV, WMV, MPEG, AVCHD
- Audio: WAV, MP3, AAC, M4A, FLAC
- Documents: PDF (police reports, witness statements)
For body-worn camera footage, FrameCounsel automatically detects metadata embedded by major manufacturers including Axon, Motorola, Digital Ally, and Getac. This metadata includes GPS coordinates, device serial numbers, and activation triggers.
Batch Import#
For cases involving multiple officers or camera angles, use File > Batch Import or press . Select an entire folder and FrameCounsel will recursively discover all supported media files, preserving the original directory structure as organizational tags.
Run Initial Analysis#
Once your evidence is imported, click the Analyze button or press . The analysis pipeline includes:
- Transcription - Speech-to-text using a fine-tuned Whisper model optimized for law enforcement audio, including handling of radio chatter, overlapping speech, and poor microphone quality.
- Speaker Diarization - Automatic identification and labeling of distinct speakers in the recording.
- Scene Detection - Frame-by-frame analysis to identify significant events, scene changes, and areas of visual interest.
- Timeline Construction - Automatic generation of a timestamped event timeline.
Analysis Speed
On an M2 Pro MacBook Pro, FrameCounsel processes approximately 10 minutes of video per minute of analysis time. Longer recordings will show a progress bar with estimated completion time.
Review Findings#
After analysis completes, FrameCounsel presents results in three synchronized panels:
- Video Player (left) - Frame-accurate playback with analysis overlays
- Transcript (center) - Time-synced transcript with speaker labels, clickable to jump to the corresponding video frame
- Findings (right) - Flagged items, contradictions, and notes
Click any finding to jump to the relevant moment in the video and transcript simultaneously. Findings are color-coded by severity: red for potential contradictions, amber for items requiring review, and blue for informational notes.
Next Steps#
Continue to the First Analysis Walkthrough for a detailed, step-by-step guide to analyzing body camera footage and comparing it against a police report.