Recap FAQ

Recap is a meeting transcription and intelligence tool that runs entirely on your device. It captures audio from your meetings (Zoom, Teams, Google Meet, or any other platform), transcribes the conversation locally, and generates structured summaries with action items.

Here's how it works:

1. Audio capture: Recap captures system audio (what you hear through your speakers/headphones) — no cloud bots joining your meetings
2. Transcription: Audio is transcribed locally using Whisper (open-source speech recognition)
3. Summarization: AI generates a summary, extracts action items, and identifies decisions
4. Organization: Transcripts and summaries are stored locally, searchable, and exportable

The key difference from other meeting tools: everything happens on your device. No audio or transcripts are sent to the cloud. Full sovereignty.

Organizations face a tension between two needs:

• Workflow automation: Cloud-based meeting tools (Otter.ai, Fireflies, Grain) are convenient but require sending your meeting audio to their servers
• Data sovereignty: Regulated industries, privacy-conscious teams, and organizations in data-residency contexts can't send sensitive conversations to third-party servers

Recap solves this by providing on-device meeting intelligence. Your audio and transcripts never leave your machine. No cloud bots silently joining calls. No transcripts sitting on a vendor's servers.

Recap is designed for regulated and privacy-conscious organizations:

• Legal firms: Attorney-client privilege requires keeping meeting conversations confidential
• Healthcare: HIPAA-adjacent workflows where patient discussions can't be sent to cloud services
• Financial services: Fintech and financial institutions with strict data controls
• Government contractors: Defense-adjacent SMBs handling sensitive government data
• EU-based organizations: GDPR data-minimization requirements
• Any organization: Where "no cloud bot joining our call" is a hard requirement

If you can't use Otter.ai or Fireflies because of data sovereignty concerns, Recap is for you.

The key difference is deployment model:

If you need meeting intelligence but can't send audio to the cloud, Recap is the only option.

Yes. Recap works with any meeting platform:

• Zoom
• Microsoft Teams
• Google Meet
• Webex
• Any other platform: If you can hear audio through your speakers/headphones, Recap can capture it

Recap captures system audio — whatever you hear through your speakers or headphones. It doesn't need to integrate with specific meeting platforms or use platform-specific bots.

No. Recap does not use cloud bots. This is a key differentiator.

Cloud-based meeting tools (Otter.ai, Fireflies, Grain) send a bot to join your meeting as a participant. The bot records the audio and sends it to the cloud for processing.

Recap works differently:

• No bot: Recap doesn't join your meeting as a participant
• System audio capture: Recap captures the audio you hear through your speakers/headphones
• Local processing: Audio is transcribed and processed on your device

Your meeting participants don't see a bot joining. Your audio never leaves your device. Full sovereignty.

Recap is both:

• Standalone: You can use Recap by itself for on-device meeting transcription and summaries
• Suite-integrated: Recap is part of the ODW.ai suite and integrates with other modules:
  • Vault: Meeting summaries and decisions auto-sync to your knowledge base
  • Loop: Action items auto-sync to your workflow automation

The free core works standalone. Suite integration (Vault/Loop sync) is part of the paid tier.

"On-device" means all processing happens on your computer:

• Audio capture: Happens on your device
• Transcription: Happens on your device (using local Whisper models)
• Summarization: Happens on your device (using local LLMs)
• Storage: Happens on your device (encrypted local storage)

No audio, transcripts, or summaries are sent to the cloud. No data leaves your machine.

This is critical for sovereignty-focused organizations. You can't claim data sovereignty if your meeting tool is sending your audio to a cloud vendor.

No. Recap is designed for non-technical users:

• Installation: Download and install like any other app (macOS, Windows, Linux)
• Setup: Configure audio source (system audio) and you're ready
• Usage: Start a meeting, Recap automatically captures and transcribes
• Review: View transcripts, summaries, and action items in the web UI

No coding, no command line, no technical expertise required. If you can use Zoom, you can use Recap.

Recap uses Whisper (OpenAI's open-source speech recognition model), which is state-of-the-art:

• Word Error Rate (WER): ≤8% on standard business English meetings (measured against human transcription baseline)
• Speaker diarization: Identifies different speakers (who said what)
• Noise handling: Handles background noise, multiple speakers, accents

Accuracy depends on:

• Audio quality: Clear audio = better accuracy
• Model size: Larger models (Whisper large-v3) are more accurate but slower
• Language: English is best supported (v1.0); other languages coming in v1.1

Whisper is OpenAI's open-source speech recognition model. It's one of the most accurate transcription models available:

• Open source: Free to use, auditable code
• Multi-language: Supports 99 languages (English is best in v1.0)
• Robust: Handles accents, background noise, multiple speakers
• Local: Runs on your device — no cloud dependency

Recap uses Whisper-family models (whisper.cpp, faster-whisper) optimized for local inference. You can choose between small (fast, less accurate), medium (balanced), or large (slow, most accurate) models.

Yes. Recap supports real-time streaming transcription:

• Streaming mode: See transcripts appear in real-time as people speak (latency ≤2 seconds)
• Batch mode: Transcribe the entire meeting after it ends (faster, more accurate)

Real-time transcription is useful for:

• Following along during the meeting
• Catching things you missed
• Live captioning for accessibility

For maximum accuracy, we recommend batch mode (transcribe after the meeting ends).

Yes. Recap includes speaker diarization (identifying who said what):

• Automatic: Recap automatically detects different speakers
• Labels: Speakers are labeled as "Speaker 1", "Speaker 2", etc.
• Manual labeling: You can rename speakers (e.g., "Speaker 1" → "John")

Speaker diarization is done locally using open-source models — no cloud processing required.

Recap captures system audio (whatever you hear through your speakers/headphones). It doesn't import audio files — it captures live audio.

Supported audio sources:

• System audio: Audio from any application (Zoom, Teams, Meet, etc.)
• Microphone: Your microphone (for capturing your own voice)
• Both: Capture system audio + microphone simultaneously

Recap doesn't support importing pre-recorded audio files (MP3, WAV, etc.) in v1.0. This is planned for v1.1.

No. Recap only captures system audio — audio from applications running on your computer (Zoom, Teams, Meet, etc.).

Recap does not support:

• Phone calls: Traditional phone calls (cellular, landline)
• SIP/VoIP: SIP-based phone systems
• Mobile apps: Recap is desktop-only (macOS, Windows, Linux)

If you need to transcribe phone calls, you'll need a dedicated phone recording solution.

Yes. Recap supports multiple export formats:

• Markdown: .md files (readable, editable)
• PDF: Formatted PDF documents
• Plain text: .txt files
• JSON: Structured data (for programmatic use)

You can export:

• Full transcript (with timestamps and speaker labels)
• Summary only
• Action items only
• All of the above

Yes. Recap includes full-text search across all past meetings:

• Search transcripts: Find specific phrases or topics
• Search summaries: Find meetings by summary content
• Search action items: Find specific action items across all meetings
• Filter by date: Narrow search by date range
• Filter by tags: Filter by custom tags or auto-generated topics

Search is local (on your device) — no cloud indexing required.

Recap competes on sovereignty + platform-agnostic design. If you need on-device processing and can't use cloud bots, Recap is the best option.

There are open-source Whisper-based tools (WhisperLive, MeetingBaize), but they have different design priorities:

DIY tools are great for experimentation, but not production-ready for business use. Recap is designed for regulated businesses that need reliability, structure, and support.

Zoom AI and Teams Copilot are platform-native meeting features, but they have limitations:

Zoom AI/Teams Copilot are convenient if you're 100% on one platform and don't have sovereignty requirements. If you use multiple platforms or need sovereignty, Recap is the better choice.

"Sovereign" means full control over your meeting data:

• Audio location: Your meeting audio stays on your device
• Processing: All transcription and summarization happens on your device
• No external dependencies: No audio sent to OpenAI, Otter.ai, or any third party
• Auditability: You can inspect the code, logs, and data flows

This is critical for regulated industries. You can't claim data sovereignty if your meeting tool is sending your audio to a cloud vendor.

Yes. Recap supports fully offline operation:

• Local models: Whisper (transcription) and LLMs (summarization) run on your device
• No internet required: Recap doesn't need internet access for transcription, summarization, or any core functionality
• Air-gapped: Deploy in air-gapped environments (no internet access at all)

This is critical for government, military, and high-security environments where internet access is restricted or prohibited.

Yes. Recap's core is open source:

• Transparent: All code is publicly available on GitHub
• Auditable: You (or your security team) can inspect the code for vulnerabilities, backdoors, or data exfiltration
• Verifiable: You can verify that Recap does what it claims — no hidden telemetry, no data sent to third parties

This is critical for sovereignty-focused organizations. You should be able to verify what your meeting tools are doing.

English only in v1.0. Multi-language support is planned for v1.1 (Q4 2026).

Whisper (the underlying transcription model) supports 99 languages, but v1.0 of Recap is optimized for English. Multi-language support will include:

• Spanish
• French
• German
• Chinese (Mandarin)
• Japanese
• And more

If you need multi-language support before v1.1, contact us — we may be able to provide early access.

Yes. Recap can handle meetings of any length:

• No time limit: Recap can transcribe meetings of any duration
• Batch processing: For long meetings, batch mode is recommended (transcribe after the meeting ends)
• Storage: Transcripts and summaries are stored locally — no cloud storage limits

For very long meetings (2+ hours), we recommend:

• Batch mode: Transcribe after the meeting ends (faster, more accurate)
• Sufficient disk space: Ensure you have enough disk space for audio and transcripts

Recap generates structured summaries:

• Executive summary: 2-3 sentence overview of the meeting
• Key topics: List of main topics discussed
• Decisions: List of decisions made during the meeting
• Action items: List of action items with assignees and deadlines (if stated)
• Key quotes: Important quotes from the meeting

Summaries are generated using local LLMs (e.g., Llama 3, Mistral) — no cloud processing required.

Recap uses AI-powered extraction to identify action items:

• Keyword detection: Identifies phrases like "I'll do X", "Can you handle Y?", "Let's follow up on Z"
• Assignee detection: Identifies who is responsible for each action item
• Deadline detection: Identifies deadlines if stated (e.g., "by Friday", "next week")
• Context understanding: Understands the context to distinguish actual action items from general discussion

Target: ≥85% precision, ≥75% recall (human-evaluated sample).

Action items are extracted locally using LLMs — no cloud processing required.

Yes. All summaries and action items are fully editable:

• Edit summaries: Modify the generated summary to add context or correct errors
• Edit action items: Add, remove, or modify action items
• Reassign: Change assignees or deadlines
• Add notes: Add your own notes to the summary

Edits are saved locally and synced with the original transcript.

Yes. Recap supports custom output templates:

• Pre-built templates: Executive summary, detailed summary, action items only, decisions only
• Custom templates: Create your own templates (e.g., focus on specific topics, include/exclude sections)
• Per-meeting templates: Use different templates for different types of meetings (e.g., standup vs. board meeting)

Templates are defined in YAML or Markdown — no coding required.

Yes. Recap supports team sharing (paid tier):

• Shared workspaces: Create shared workspaces for your team
• Access control: Control who can view/edit meeting summaries
• Notifications: Notify team members when new summaries are available
• Comments: Team members can add comments to summaries

For the free tier, summaries are local-only (single user). Team features require the paid tier.

Yes. Recap supports action item export:

• Manual export: Export action items as CSV, JSON, or Markdown
• Loop sync: Auto-sync action items to ODW.ai Loop (workflow automation) — paid tier
• Custom integrations: Use the API to integrate with your task management tool (Jira, Asana, Trello, etc.)

Loop sync automatically creates tasks, assigns them, and sets deadlines based on the extracted action items.

Yes. Recap includes decision search:

• Search decisions: Find specific decisions across all past meetings
• Filter by date: Narrow search by date range
• Filter by topic: Filter by meeting topic or custom tags
• View context: Click on a decision to see the full meeting transcript and context

This is useful for tracking what was decided, when, and by whom — critical for accountability and audit trails.

Summary quality depends on:

• Transcription accuracy: Better transcription = better summaries
• LLM quality: Better LLMs (e.g., Llama 3 70B) generate better summaries
• Meeting clarity: Clear, structured meetings = better summaries

Target: ≥4.0/5.0 in user satisfaction surveys.

Summaries are AI-generated, so they may not be perfect. We recommend reviewing and editing summaries for important meetings.

Yes. Recap is built with security best practices:

• On-device processing: All audio, transcripts, and summaries stay on your device
• Encryption: AES-256 encryption for local storage
• Access control: Password protection, role-based access (paid tier)
• Audit logging: Tamper-evident logs of all access (paid tier)
• Open source: Code is auditable — you can verify there are no backdoors or data exfiltration

Recap is designed for regulated industries. If you can use Recap, you can pass a compliance audit.

No. Recap runs entirely on your device. ODW.ai has zero access to your audio, transcripts, or summaries.

Unlike cloud-based meeting tools (Otter.ai, Fireflies), Recap doesn't send your audio to a third party. Everything stays on your machine.

This is critical for sovereignty-focused organizations. You can't claim data sovereignty if your meeting tool is sending your audio to a cloud vendor.

Yes. Recap is HIPAA-compatible:

• On-device processing: PHI stays on your device — no data sent to third parties
• Encryption: AES-256 encryption for local storage
• Access control: Password protection, role-based access
• Audit logging: Tamper-evident logs of all access

Recap helps you meet the technical safeguards required by HIPAA. You still need to address administrative and physical safeguards separately.

Yes. Recap is GDPR-compatible:

• Data residency: All data stays on your device (or your infrastructure for team deployments)
• Right to erasure: Delete meetings and associated data on request
• Right to access: Export meeting data on request
• Data minimization: No data sent to third parties

Recap helps you meet the technical and organizational measures required by GDPR (Article 32). You still need to address legal and procedural requirements separately.

Yes. Recap supports air-gapped deployment (no internet access):

• Offline installation: Install Recap without internet access
• Local models: All models (Whisper, LLMs) run on your device — no API calls to external services
• No external dependencies: Recap doesn't require internet access for transcription, summarization, or any core functionality

Air-gapped deployment is common for government contractors, military, and high-security environments.

Recap provides access control (paid tier):

• Password protection: Require a password to access Recap
• Role-based access: Control who can view/edit meeting summaries
• Audit logging: All access is logged (who accessed what, when)
• Alerting: Get notified of unauthorized access attempts

For the free tier, Recap is single-user (local only). If someone has physical access to your device and your password, they can access your data — so use strong passwords and encryption.

Yes. Recap supports retention policies (paid tier):

• Auto-delete: Automatically delete meetings after N days (e.g., 30 days, 90 days, 1 year)
• Per-workspace policies: Different retention policies for different workspaces
• Manual override: Mark specific meetings as "do not delete"

This is useful for compliance (e.g., GDPR data minimization) and storage management.

Yes. Recap's core is open source:

• Transparent: All code is publicly available on GitHub
• Auditable: You (or your security team) can inspect the code for vulnerabilities, backdoors, or data exfiltration
• Verifiable: You can verify that Recap does what it claims — no hidden telemetry, no data sent to third parties

This is critical for sovereignty-focused organizations. You should be able to verify what your meeting tools are doing.

Recap uses open-source models for transcription and summarization:

Transcription:

• Whisper: OpenAI's open-source speech recognition model (whisper.cpp, faster-whisper)
• Model sizes: Small (fast, less accurate), Medium (balanced), Large (slow, most accurate)

Summarization:

• Llama 3: Meta's open-source LLM (8B, 70B)
• Mistral: Mistral AI's open-source LLM
• Other: Any model compatible with Ollama or llama.cpp

You can choose which models to use based on your priorities (speed vs. accuracy vs. resource usage).

Not in v1.0. Recap is designed for on-device processing — no cloud API support in v1.0.

This is by design. Recap's core promise is sovereignty — no data leaves your device. Adding API support would break that promise.

If you need API-based models (for higher quality), you can:

• Use local models: Llama 3 70B, Mistral Large, etc. — quality is approaching API models
• Wait for v2.0: API support is planned for v2.0 (Q2 2027) with strict opt-in and data controls

Choosing a Whisper model depends on your priorities:

Recommendation:

• Real-time transcription: Use Whisper small or medium (fast enough for streaming)
• Batch transcription: Use Whisper large-v3 (most accurate, no time pressure)
• Low-resource devices: Use Whisper small (runs on older hardware)

Choosing a summarization model depends on your priorities:

Recommendation:

• Fast summaries: Use Llama 3 8B or Mistral 7B (good quality, fast)
• Best quality: Use Llama 3 70B (best quality, but slow and resource-intensive)
• Low-resource devices: Use Llama 3 8B or Mistral 7B (runs on older hardware)

Yes. Recap is model-agnostic — you can swap models at any time:

• No code changes: Just update the model configuration in the settings
• Per-meeting: Use different models for different meetings (e.g., small model for standups, large model for board meetings)
• Download models: Download new models via the UI (one-click)

This future-proofs your investment — when better models are released, you can adopt them immediately without re-architecting your system.

Yes. Recap is model-agnostic — you can use any compatible model:

• Whisper-family: Any Whisper variant (whisper.cpp, faster-whisper, distil-whisper, etc.)
• LLM-family: Any LLM compatible with Ollama or llama.cpp (Llama 3, Mistral, Gemma, Qwen, etc.)
• Custom models: If you have a custom model, you can integrate it via the API

When better models are released, you can adopt them immediately — no code changes, no vendor lock-in.

Not in Recap v1.0. Recap uses existing models — it doesn't support fine-tuning.

However, you can:

• Use custom prompts: Customize the summarization prompt to guide the model's behavior (e.g., "Focus on legal decisions" or "Extract action items for the engineering team")
• Fine-tune externally: Fine-tune a model outside of Recap (using tools like Hugging Face Transformers), then use the fine-tuned model in Recap via Ollama or llama.cpp

Fine-tuning support is planned for v2.0 (Q2 2027).

Recap is available for macOS, Windows, and Linux:

• macOS: Download the .dmg file from odw.ai/recap, drag to Applications
• Windows: Download the .exe installer from odw.ai/recap, run the installer
• Linux: Download the .deb (Ubuntu/Debian) or .rpm (Fedora/RHEL) package, install via package manager

Installation takes <2 minutes. No technical skills required.

Recap's requirements depend on your usage:

For real-time transcription, we recommend a modern CPU (M1/M2/M3 for macOS, Intel i7+ or AMD Ryzen 7+ for Windows/Linux). GPU is optional but speeds up inference.

No. Recap is designed for non-technical users:

• Installation: Download and install like any other app
• Setup: Configure audio source (system audio) and you're ready
• Usage: Start a meeting, Recap automatically captures and transcribes

No coding, no command line, no technical expertise required. If you can use Zoom, you can use Recap.

Yes. Recap supports team deployments (paid tier):

• Self-hosted server: Deploy Recap on a central server (Docker-based) for team access
• Shared workspaces: Create shared workspaces for your team
• Access control: Control who can view/edit meeting summaries
• Admin dashboard: Monitor usage, manage users, configure settings

Team deployments require some technical setup (Docker, networking). We provide deployment guides and support.

Recap captures system audio — whatever you hear through your speakers/headphones:

1. Open Recap settings: Click the gear icon in the Recap UI
2. Select audio source: Choose "System Audio" (captures all audio from your computer)
3. Optional — Microphone: If you want to capture your own voice, enable "Microphone" as well
4. Test: Click "Test Audio" to verify Recap is capturing audio correctly

That's it. Recap will automatically capture audio from any meeting platform (Zoom, Teams, Meet, etc.).

Yes. Recap offers a free core tier:

• No time limit: Use the free tier forever
• Full features: On-device transcription, local summaries, action item extraction
• No credit card: No payment information required

The free tier is perfect for testing Recap with your meetings. If you like it, you can upgrade to the paid tier for team features and suite integration.

Recap updates automatically:

• Desktop app: Recap checks for updates on startup and prompts you to install
• Manual update: You can also check for updates manually via the settings menu
• Models: Model updates are separate — you can update models via the model manager in the UI

Updates are typically small and fast (<5 minutes to install).

Yes. You can install Recap on multiple devices:

• Free tier: Single-user, local-only (each device has its own data)
• Paid tier: Sync data across devices (optional)
• Team tier: Centralized server, multiple users access the same data

For the free tier, each device operates independently — no sync between devices. For the paid tier, you can optionally sync data across your devices.

Recap pricing is based on features and deployment:

Note: Recap is not sold standalone for enterprise — it's part of the ODW.ai suite bundle. Contact us for suite pricing.

It depends. Otter.ai and Fireflies cost $10-$30/user/month. For a 10-person team, that's $100-$300/month.

Recap costs $0-$199/month (flat rate, not per user for the free core). For a 10-person team:

• Free tier: $0 (but single-user per device, no sync)
• Pro tier: $49/month (team workspaces, shared meetings)
• Enterprise tier: $199/month (suite integration, audit logging)

For small teams, Recap is cheaper. For large teams, Otter.ai/Fireflies may be cheaper per user — but Recap offers sovereignty, which they don't.

No hidden costs. The total cost of ownership includes:

• Recap subscription: Monthly fee based on tier (or $0 for free core)
• Infrastructure: You pay for your own devices (if self-hosting)

That's it. No setup fees, no surprise charges, no per-user fees (for the free core).

Yes. The free core tier is 100% free:

• No time limit: Use it forever
• No feature limits: On-device transcription, local summaries, action items
• No user limits: Install on as many devices as you want (single-user per device)

The only limitations:

• Team features: Shared workspaces, team sync, audit logging are paid
• Suite integration: Vault/Loop sync is paid (Enterprise tier or suite bundle)
• Support: Community support only (GitHub issues, Discord). No email or priority support.

We believe in open source. If you don't need team features or suite integration, you can use Recap for free forever.

We accept:

• Credit/debit cards: Visa, Mastercard, American Express (via Stripe)
• Bank transfer: For annual subscriptions or enterprise customers
• Cryptocurrency: Bitcoin, Ethereum, USDC (for privacy-conscious customers)

Invoicing is available for annual subscriptions. Contact us for details.

Yes. We offer a 30-day money-back guarantee. If you're not satisfied within the first 30 days, contact us for a full refund — no questions asked.

After 30 days, you can cancel anytime. You'll continue to have access until the end of your billing period, but no refund is provided for partial months.

Yes. Recap is part of the ODW.ai suite. You can:

• Purchase Recap standalone: Use Recap by itself (free core or paid tier)
• Purchase the full suite: Get Desk, Voice, Vault, Recap, Shield, and other modules at a discounted price

The suite bundle includes Recap at a 20-30% discount compared to purchasing separately. And all modules integrate natively — Recap feeds meeting intelligence into Vault (knowledge base) and Loop (workflow automation).

Recap can handle unlimited meetings:

• No meeting limit: Transcribe as many meetings as you want
• No time limit: Transcribe meetings of any duration
• Storage: Limited only by your disk space (typical meeting = 50-200 MB)

For large volumes (100+ meetings per week), we recommend:

• Sufficient disk space: Ensure you have enough storage for audio and transcripts
• Batch mode: Transcribe after meetings end (faster, more accurate)
• Team deployment: Use a central server for team access (paid tier)

Recap supports unlimited users:

• Free tier: Single-user per device (each user has their own local data)
• Pro tier: Team workspaces with multiple users
• Enterprise tier: Centralized server, unlimited users

For team deployments, Recap scales horizontally — add more server resources to handle more users.

If Recap crashes, your audio is not lost:

• Audio buffer: Recap buffers audio locally — if it crashes, the buffered audio is saved
• Auto-restart: Recap automatically restarts and resumes capturing audio
• Manual recovery: You can manually recover the buffered audio and transcribe it later

For critical meetings, we recommend:

• Test before: Test Recap before important meetings
• Backup recording: Use a backup recording method (e.g., Zoom's built-in recording) for critical meetings

Recap includes a performance dashboard:

• System health: CPU, memory, disk usage
• Transcription metrics: Number of meetings transcribed, average transcription time, accuracy estimates
• Model metrics: Which models are being used, resource usage
• Storage metrics: Disk space used by meetings, transcripts, and models

For team deployments, the admin dashboard provides additional metrics (user activity, workspace usage, etc.).

Recap data is stored locally on your device. Back up using standard backup tools:

• macOS: Time Machine, or manually copy the Recap data folder
• Windows: File History, or manually copy the Recap data folder
• Linux: rsync, or manually copy the Recap data folder

Recap data location:

• macOS: ~/Library/Application Support/Recap/
• Windows: %APPDATA%\Recap\
• Linux: ~/.config/Recap/

We recommend backing up this folder regularly (e.g., daily via Time Machine or cloud sync).

Yes. Recap supports team workspaces (paid tier):

• Separate workspaces: Create separate workspaces for different teams or departments
• Access control: Control who can access which workspaces
• Isolation: Each workspace has its own meetings, summaries, and settings

This is useful for organizations with strict data separation requirements or multi-team deployments.

Recap is designed to be intuitive, but we provide training resources:

• Documentation: Comprehensive user guides for admins and end users
• Video tutorials: Step-by-step walkthroughs of key features
• Onboarding session: For paid customers, we offer a 1-hour onboarding session (remote)
• Support: Priority support for Pro and Enterprise customers

Most teams are productive within 30 minutes of training. The UI is particularly intuitive — no technical skills required for end users.

Recap integrates natively with other ODW.ai products (paid tier):

• Vault: Meeting summaries and decisions auto-sync to your knowledge base
• Loop: Action items auto-sync to your workflow automation (create tasks, assign, set deadlines)
• Shield: Recap emits audit logs to Shield for governance and compliance

If you're using the ODW.ai suite, Recap integration is automatic — no separate configuration required.

Yes. Recap can auto-sync meeting summaries to Vault (Enterprise tier or suite bundle):

• Auto-sync: Meeting summaries and decisions are automatically indexed in Vault
• Searchable: Meeting content becomes searchable in Vault alongside other documents
• Bi-directional: Vault entries can reference source meetings
• Configurable: Choose which meetings sync, which tags trigger sync, etc.

This turns your meetings into a searchable knowledge base — no more losing important decisions in meeting notes.

Yes. Recap can auto-sync action items to Loop (Enterprise tier or suite bundle):

• Auto-create tasks: Action items are automatically created as tasks in Loop
• Assignees: Tasks are assigned to the correct people (detected from the meeting)
• Deadlines: Deadlines are set based on what was stated in the meeting
• Workflows: Action items can trigger workflows (e.g., "When action item is created → notify assignee → set reminder")

This turns your meetings into actionable workflows — no more manual task creation after meetings.

Not in v1.0. Recap doesn't integrate with calendars in v1.0.

Recap captures system audio regardless of calendar — it doesn't need to know about your meetings in advance. You just start a meeting, and Recap automatically captures and transcribes.

Calendar integration (auto-start transcription based on calendar events) is planned for v1.1 (Q4 2026).

Not directly. Recap doesn't post to Slack or Teams channels in v1.0.

However, you can:

• Export and share: Export summaries as Markdown or PDF, then share via Slack/Teams
• Loop integration: If you're using ODW.ai Loop, you can configure Loop to post summaries to Slack/Teams
• Custom integration: Use the API to build a custom integration with Slack/Teams

Native Slack/Teams integration is planned for v1.2 (Q1 2027).

Not in v1.0. Recap doesn't provide a public API in v1.0.

API access is planned for v1.2 (Q1 2027). The API will allow:

• Query meetings: Retrieve meeting transcripts, summaries, and action items
• Trigger transcription: Programmatically transcribe audio files
• Integrate with custom tools: Build custom integrations with Recap

In the meantime, you can use the export features (Markdown, JSON) to integrate with other tools.

Yes. Recap can be used with existing AI tools:

• Export transcripts: Export transcripts as JSON, then use them in your AI pipelines
• Export summaries: Export summaries as Markdown, then use them in your knowledge base
• Custom integrations: Use the export features to integrate with LangChain, LlamaIndex, etc.

Recap is designed to be a meeting intelligence layer — it provides transcription and summarization, and you can use the output in your existing workflows.

Recap is powerful, but it has limitations in v1.0:

• English only: Multi-language support planned for v1.1 (Q4 2026)
• No video recording: Audio-only (video adds significant complexity)
• No real-time translation: Transcription only; translation is a future consideration
• No mobile apps: Desktop-only for v1.0 (meetings happen on desktop)
• No phone call recording: Only system audio (no phone call recording)
• No public API: API planned for v1.2 (Q1 2027)
• No calendar integration: Planned for v1.1 (Q4 2026)
• No cloud API support: On-device only in v1.0 (API support planned for v2.0)

Yes. Multi-language support is planned for v1.1 (Q4 2026):

• Spanish
• French
• German
• Chinese (Mandarin)
• Japanese
• And more

Whisper (the underlying transcription model) already supports 99 languages — v1.0 of Recap is just optimized for English. v1.1 will expand language support.

Not in the current roadmap. Video recording adds significant complexity (storage, processing, UI) and is not needed for transcription value.

Recap is focused on audio transcription and meeting intelligence. If you need video recording, we recommend using your meeting platform's built-in recording (Zoom, Teams, Meet) or a dedicated video recording tool.

If video recording is a must-have for your use case, contact us — we may reconsider based on customer demand.

Planned for v2.0 (Q2 2027). Real-time translation will include:

• Live translation: Translate speech in real-time (e.g., someone speaks English, you see Spanish subtitles)
• Post-meeting translation: Translate transcripts and summaries after the meeting
• Multi-language support: Translate between 50+ languages

This is a high-priority feature for global teams and multilingual organizations.

Key milestones:

• v1.0 (Q3 2026): MVP launch. On-device transcription, local summaries, action items, macOS/Windows/Linux support.
• v1.1 (Q4 2026): Multi-language support, calendar integration, improved speaker diarization.
• v1.2 (Q1 2027): Public API, Slack/Teams integration, advanced analytics.
• v2.0 (Q2 2027): Real-time translation, cloud API support (opt-in), model fine-tuning.

Roadmap is subject to change based on customer feedback. Contact us for the latest roadmap or to request features.

Recap is not the right fit if:

• You don't have sovereignty requirements: If you don't care about data sovereignty, cloud-based tools (Otter.ai, Fireflies) are easier (no local processing).
• You need video recording: Recap is audio-only. If you need video, use your meeting platform's built-in recording.
• You need phone call recording: Recap only captures system audio. If you need phone call recording, use a dedicated phone recording solution.
• You need mobile apps: Recap is desktop-only. If you need mobile transcription, use a mobile-first tool.
• You need real-time translation: Recap doesn't support translation in v1.0. If you need translation, wait for v2.0 or use a dedicated translation tool.

No. No AI transcription system can guarantee 100% accuracy. Recap's accuracy depends on:

• Audio quality: Clear audio = better accuracy
• Model size: Larger models (Whisper large-v3) are more accurate
• Speaker clarity: Clear speakers = better accuracy
• Background noise: Less noise = better accuracy

Target: ≤8% Word Error Rate (WER) on standard business English meetings. This means 92%+ accuracy.

For critical meetings (legal, medical), we recommend human review of the transcript.

Three paths:

1. Free download: Download Recap from odw.ai/recap. No credit card required. Try it with your meetings and see if it works for your use case.
2. Demo: Book a demo with our team. We'll walk you through Recap, answer your questions, and help you evaluate if it's right for your business.
3. Contact sales: For enterprise customers or suite bundles, contact our sales team for a tailored quote.