Najva
Najva is a free macOS menubar app that does offline Whisper speech-to-text, then optionally hands your transcript to Claude, GPT, or Gemini using your own API
If you want dictation on a Mac that keeps working with the Wi-Fi off, Najva is one of the few options that does not watermark or meter anything — it is listed at ₹0+ on Gumroad with 30 ratings at 5.0, and transcription runs on-device through Whisper. The catch is that the intelligence layer is your problem: you supply API keys, and providers like Anthropic, OpenAI, and Google bill you separately. If you never configure one, you are left with plain but solid transcription. Named alternatives sit in different shapes — Superwhisper and MacWhisper for Mac dictation, and cloud subscriptions such as Otter for teams. Solo use only; there is nothing here for a shared workspace.
Verified 3d ago · liveness 58/100 · cite: rightaichoice.com/tools/najva
- Solo macOS users who want offline dictation that works without internet
- Writers and developers who dictate drafts and want AI to rewrite or expand them
- Privacy-conscious professionals, including clinicians, who cannot send audio to a cloud service
- Students capturing lectures or notes using their own AI provider key
- Teams needing shared transcription workspaces or collaborative editing
- Users who want one subscription covering both transcription and AI in a single bill
- People who need a Windows or Linux build
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Najva if you need a shared team workspace, a single invoice covering both transcription and AI, or a Windows/Linux build — it is a solo Mac dictation app and every team-shaped requirement falls outside it.
AI processing is billed by your provider, not by Najva — every request you send to Claude, GPT, Gemini, OpenRouter, or Groq lands on that provider's invoice separately.
Najva itself is listed at ₹0+ / "Forever Free" on Gumroad — the developer states there are no premium tiers or feature gates. Your real spend is your AI provider: Claude, OpenAI, Gemini, OpenRouter, or Groq bill you directly per request, and Ollama or vLLM cost nothing beyond your own hardware. Against a cloud transcription subscription that bundles both, Najva separates the free app from a metered provider bill only you control.
In short
Najva — Najva is a free macOS menubar app that does offline Whisper speech-to-text, then optionally hands your transcript to Claude, GPT, or Gemini using your own API. Best for Solo macOS users who want offline dictation that works without internet, Writers and developers who dictate drafts and want AI to rewrite or expand them, Privacy-conscious professionals, including clinicians, who cannot send audio to a cloud service. Free to use.
Viability Score
How well maintained and how widely used is Najva? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Offline speech-to-text transcription powered by Whisper, running on-device
- Real-time dictation with Fast Mode pasting processed text into the active field
- Context awareness that automatically includes your selected text as context
- Connect Anthropic Claude (listing names Claude 3.7)
- Connect OpenAI GPT models including GPT-4.5, o3, and o4-mini
- Connect Google Gemini 2.5 Pro and Flash models
- Connect OpenRouter, Groq, Ollama, or vLLM
- Bring your own API key for full control over provider and spend
- Local model support via Ollama and vLLM for fully on-device AI processing
- Transcribe existing audio files, not just live microphone input
- Smart Screenshots capture from screen, FaceTime camera, or iPhone camera
- Use an iPhone as a remote microphone for the Mac app
- Multiple language speech recognition
- Native Swift macOS app using under 90MB of RAM when idle
- Zero trackers, zero analytics, no data collection
About Najva
Najva (Persian for "whisper") is a native Swift macOS menubar app that turns speech into text on-device using Whisper, so dictation keeps working with Wi-Fi off and your audio never leaves the Mac unless you choose to send it onward. The AI layer is bring-your-own-key: you can connect Anthropic Claude (the listing names Claude 3.7), OpenAI's GPT line (GPT-4.5, o3, o4-mini), Google Gemini 2.5 Pro and Flash, OpenRouter for open-source models, Groq for fast responses, and Ollama or vLLM for fully local processing. The practical hook is context awareness — Najva automatically attaches whatever text you have selected and passes it to the model, so a spoken command like "summarize this" arrives with the paragraph it refers to. Fast Mode pastes the processed result straight into the active text field, Smart Screenshots pull visuals in from your screen, the FaceTime camera, or an iPhone camera, an iPhone can act as a remote microphone, and existing audio files can be transcribed after the fact. It is built by Bardia Khosravi, a medical doctor and programmer with over a decade of full-stack experience, and the design leans toward settings where accuracy and privacy matter more than collaboration. The app is not metered by the vendor — it is listed at ₹0+ / "Forever Free" on Gumroad, with 30 ratings and a 5.0 average, and the only bill is whatever your chosen AI provider charges for the requests you send it. Teams, Windows and Linux users, and anyone wanting a single subscription covering both transcription and AI are outside what it does.
Behind the Verdict
Najva's proposition is narrow and it is honest about that. Speech recognition runs locally through Whisper, so the app keeps functioning on a plane and your audio stays on the Mac by default. The AI layer is entirely opt-in and entirely yours: you bring an API key for Claude (the listing cites Claude 3.7), OpenAI's GPT-4.5/o3/o4-mini, Gemini 2.5 Pro or Flash, OpenRouter, Groq, or local models through Ollama and vLLM. Because nothing is metered by the developer, the only variable cost is what your provider charges. Where Najva earns its place is context awareness — the app automatically includes your currently selected text as context before it sends a request, so you can highlight a paragraph, say "summarize this," and the instruction carries the source material with it. Fast Mode pastes the finished text into whatever field has focus, Smart Screenshots pull in visuals from the screen, the FaceTime camera, or an iPhone camera, an iPhone can act as a remote microphone, and pre-recorded audio files can be transcribed after the fact. The engineering choices are visible in the numbers the developer publishes: a native Swift app that idles under 90MB of RAM, with no trackers, no analytics, and no data collection. The weaknesses follow from the same design. There is no collaboration surface, no shared workspace, no cloud sync — a team cannot work in it together. The vendor's curation of AI models is whatever you configure; Najva does not arbitrate model quality for you, and every model you pick carries its own cost and its own privacy posture once you send text out. The build is macOS-only, so Windows and Linux are simply out of scope. And because it is a solo-developer free app rather than a company product, the things enterprises ask for — SLAs, support contracts, a roadmap commitment — are not part of the deal. That is a fair trade for a solo Mac user who already pays for a Claude, OpenAI, or Gemini key; it is the wrong tool for a clinic or a team that needs one invoice, one admin console, and one throat to choke.
Researching Najva? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Najva actually fits — and what changes day-one when you adopt it.
You open a draft, highlight the paragraph you just wrote, and speak "rewrite this tighter" into the menubar shortcut with your Claude key configured and Fast Mode on.
Outcome: Najva sends the instruction plus your selected paragraph as context, and Fast Mode pastes the rewritten text straight back into the active field — no window switching.
You dictate notes with Wi-Fi deliberately off, keeping Whisper transcription on-device, and only turn on a local Ollama model when you want text processed further.
Outcome: Speech never leaves the Mac, AI processing stays on the machine, and no audio or text reaches a hosted provider.
You set your iPhone as a remote microphone feeding the Mac app, capture slides with Smart Screenshots from the phone camera, and dictate over the top of them.
Outcome: One app collects both the spoken notes and the visual context, and you can transcribe the recordings afterwards if you missed something live.
Use Cases
- Dictate notes into any app from the menubar without switching windows.
- Highlight a paragraph, say "summarize this," and let selected-text context carry the source material.
- Capture a screenshot from screen, FaceTime camera, or iPhone camera and dictate context in one step.
- Transcribe voice memos or existing audio files offline with no internet connection.
- Draft emails or documents hands-free, using an AI provider to expand short spoken phrases.
- Capture meeting action items without opening a separate note-taking app.
- Run AI processing entirely locally through Ollama or vLLM when the text cannot leave the machine.
Models Under the Hood
as of 2026-09-01
Limitations
- Najva is macOS-only, so Windows and Linux are out of scope.
- Core transcription is offline through Whisper, but accuracy for heavy accents or niche dialects may suffer without an internet connection.
- There is no collaboration surface, no team workspace, and no cloud sync.
- The developer documents the app rather sparsely, so troubleshooting leans on the listing page and community comments rather than a help center.
- Because it is a free solo-developer app, there is no enterprise support or SLA, and you should not expect vendor-backed response times.
- The AI layer depends on third parties: you create accounts and keys with Anthropic, OpenAI, Google, OpenRouter, Groq, or run Ollama/vLLM locally, and anything you send to a hosted provider leaves your Mac.
- If you need high-volume transcription with editing, formatting, or a multi-device workflow, this is not the fit.
as of 2026-10-08
Verification history
We have re-verified Najva 11 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
Showing the 6 most recent of 11 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Najva tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Forever Free
₹0+
Ideal for
Solo Mac users who want offline Whisper dictation and are willing to bring their own Claude, OpenAI, Gemini, OpenRouter, or Groq key — or run Ollama/vLLM locally.
What this tier adds
Starting tier — the developer lists it at ₹0+ with no premium features or feature gates; AI provider usage is billed separately by whichever provider you connect.
Where the pricing makes sense
The company stage and team size where Najva's pricing actually pencils out — and where peers do it cheaper.
Najva itself is listed at ₹0+ / "Forever Free" on Gumroad — the developer states there are no premium tiers or feature gates. Your real spend is your AI provider: Claude, OpenAI, Gemini, OpenRouter, or Groq bill you directly per request, and Ollama or vLLM cost nothing beyond your own hardware. Against a cloud transcription subscription that bundles both, Najva separates the free app from a metered provider bill only you control.
Setup time & first value
How long it actually takes to get something useful out of Najva — broken out by persona, not the marketing-page minute.
Solo Mac user: install from the Gumroad listing, pick your microphone, and you can dictate offline in roughly five minutes — no key required. Adding AI takes another five: paste an API key for Claude, OpenAI, Gemini, OpenRouter, or Groq and the processing layer is live. Local-model users running Ollama or vLLM should budget longer, since the model itself has to be pulled and configured before
Switching to or from Najva
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From typing-only workflow: install Najva, keep the menubar shortcut, and dictate into the same apps you already use — nothing else in your setup changes.
- →From a cloud transcription subscription: move live dictation to Najva's on-device Whisper and bring your existing Claude, OpenAI, or Gemini key to keep the AI processing you were paying for.
- →From macOS built-in Dictation: switch the menubar shortcut, then optionally add an API key so spoken commands get summarized or rewritten instead of only transcribed.
- ↗To a team transcription platform: export your notes and text and move to a shared workspace tool, since Najva has no multi-user surface to hand off.
- ↗To a bundled cloud subscription: if you would rather pay one vendor for transcription and AI together, migrate your provider usage into a service that bills both on a single plan.
Integrations
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “Najva”, and we withheld 6: 6 could not be judged, because “Najva” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Najva.
Official links
Tools that pair well with Najva
Common stack mates teams adopt alongside Najva, with the specific reason each pairing earns its keep.
TalkNotes
TalkNotes is a voice-to-text note app that records up to 2 hours per note and reshapes your speech into 100+ structured output styles.
Harker
Harker is a free macOS voice-to-text app that transcribes speech on-device and pastes it into whatever app you're already typing in.
Letterly
Voice dictation that turns speech into emails, notes, and meeting recaps across iPhone, Android, Mac, Windows, and Web.
Featured Head-to-Head Comparisons
Najva vs Gem
Gem and Najva serve completely different needs: Gem is a full-blown AI recruiting platform for teams filling roles at scale, while Najva is a free, privacy-focused dictation tool for individual macOS users. Unless you need both enterprise recruiting and offline speech-to-text, choose based on your job function. For most buyers, the comparison is irrelevant — pick the tool that matches your workflow.
Najva vs Guesty
For individual privacy-conscious users needing offline dictation and AI text enhancement, Najva's free macOS app is perfect. For vacation rental pros managing dozens of listings across channels, Guesty’s AI-powered automation and recent bank reconciliation add serious operational value. They serve entirely different needs.
Najva vs Poke Interaction Co
Choose Najva if you need offline, privacy-first speech-to-text on macOS with AI text processing and work primarily on your desktop. Choose Poke if you want a proactive AI assistant that handles email, calendar, and automations from within your messaging apps (iMessage, WhatsApp, Telegram) and are okay with cloud processing. There is minimal overlap; Najva is a dictation tool, Poke is a life-management chatbot.
Assemblyai vs Najva
Choose Najva if you're a solo macOS user needing free, offline dictation and privacy. Choose AssemblyAI if you're a developer building scalable voice applications requiring real-time streaming, 99-language support, and advanced speech understanding — AssemblyAI is a production-ready API platform, not a desktop app. There's no direct overlap; pick based on your deployment needs: local vs. cloud, free vs. pay-per-use.
Alternatives to Najva
View all
TalkNotes
TalkNotes is a voice-to-text note app that records up to 2 hours per note and reshapes your speech into 100+ structured output styles.
Frequently Asked Questions
Categories
Best-of guides
Used Najva? Help shape our editorial sentiment research.