# Tenvar > Tenvar is a private desktop workspace for research, transcription, notes and writing. It brings documents, recordings and imported notes into projects, uses on-device AI for transcription and cited questions, and turns selected evidence into notes, manuscripts and reports. Version: 0.4.0. Product facts checked 2026-09-30. ## Overview - Website: https://tenvar.app/ - Capabilities: https://tenvar.app/#capabilities - Formats: https://tenvar.app/#formats - Privacy: https://tenvar.app/#privacy - Pricing: https://tenvar.app/#pricing - Downloads and requirements: https://tenvar.app/#download - Published release: https://github.com/jBlanca/tenvar.app/releases/tag/v0.4.0 ## Capabilities ### Turn recordings into working material. Transcribe audio and video on your computer. Review the text beside playback, correct it, search it, and keep useful moments as marks. Speech recognition, speaker identification and language-model tools are separate steps, so you choose the processing you need. - Speaker labels help distinguish voices; you can name and remember speakers locally. - Whisper and Parakeet support word timings. Timing detail and language coverage depend on the selected speech model. - There is no subscription minute allowance or per-file transcription charge. Processing speed depends on your hardware and model. ### Keep the conversation and its context. Record your microphone and system audio, with supported options for meeting-window capture, camera video and picture-in-picture. Review a recording after it ends and transcribe it into the same source workspace used for imported media. - Capture depends on operating-system permissions, available devices and the chosen audio source. - Recording starts when you start it. Your meeting content is processed locally. - Version 0.4.0 transcribes after recording; it does not provide live meeting captions. ### Read the page. Work with the text. Read PDFs, books, Office documents, spreadsheets, emails and images alongside their extracted text. Use local OCR for scans and image-based pages, then search, ask questions and mark passages with the original in reach. - Page and region references keep document evidence connected to its source. - Some legacy Office formats require the optional LibreOffice document runtime, downloaded on demand. - Extraction and OCR are subject to layout, scan quality and format limitations. Tenvar is not a replacement for editing an original spreadsheet or slide deck. ### Ask your material a better question. Search by meaning and ask questions about a file, a project or the material gathered in a thread. On-device models retrieve relevant source passages and produce answers with references you can open to inspect a page, passage or recorded moment. - Keep useful answers and source-derived results in the project that produced them. - Local tools can summarize, clean up or extract information from a source; availability depends on the selected model and source. - A citation is a route back to evidence, not a guarantee that an AI answer is correct. Check the original before relying on it. ### Connect the evidence you choose. Highlight or underline passages, add comments, and keep page regions, pictures and recorded moments as marks. Gather marks from different files into a thread around a question, theme or argument. - Threads bring evidence from recordings, books, documents and your own notes into one place. - The canvas and timeline help you navigate sources, their marks and the connections you made. - Connections reflect your filing and recorded relationships. Tenvar does not invent a graph of supposed connections between your ideas. ### Write with your references beside you. Create editable notes with rich text, headings, lists, tables, pictures and source links. Reference another source or insert selected evidence while keeping the Inspector available for your material. Saved notes can be searched, questioned and cited like other sources. - Your own writing and imported notes share the same editing workflow. - Use notes for research, reading, planning and drafting within a project. - Export your work in usable formats. The 0.4.0 release does not continuously synchronize notes into an external Markdown folder. ### Give longer writing an outline. Arrange notes into manuscripts with parts, chapters and scenes. Reorder the outline, keep synopsis and viewpoint information, and park material outside the reading order without discarding its note. - A note can belong to more than one manuscript in the same project; its writing stays shared while each book keeps its own order. - Read through a saved draft, preserve named snapshots, and restore a draft as a separate manuscript. - Hand off a manuscript as DOCX or PDF. The export identifies material, such as playable media, that cannot travel in those formats. ### Send the evidence with the result. Write a flowing document or arrange designed report pages using your selected material. Place quotations, page images and media evidence beside the argument they support. Export a self-contained HTML report with playable audio and video excerpts. - The HTML report opens in a browser without Tenvar or an internet connection. - PDF export provides a printable document. Playable audio and video belong to the HTML handoff, not the PDF. - You compose and review the report. The experimental report-wide AI editing assistant is not included in 0.4.0. ### Take the transcript back to the video. Review and edit captions against playback. Export subtitle files for another editor, embed a subtitle track in supported media containers, or burn captions into a video using the bundled media tools. - Subtitle exports include SRT, WebVTT, ASS, SBV and plain text. - Timing quality depends on the speech engine, source audio and your edits. - Use this for recorded material; the subtitle workflow is separate from live meeting captions. ### Find material before bringing it in. Shelves catalogs recognized files inside the folders you choose to cover. Browse and inspect that material in place, then use Add as source when a file belongs in a project. Tenvar keeps its own project copy and leaves the original alone. - Bring recordings, documents, pictures and notes into the same project workflow. - Identical compatible sources can be reused across unencrypted projects rather than storing another copy. - Protected projects keep their own protection boundaries. Browsing your selected folders does not make their contents a cloud library. ### Keep references you can return to. Import a web page into a local reading and source workflow, or bring in saved email files and mail archives. Their extracted content can join the same search, marks and cited-question workflow as your other documents. - Online web and email ingestion requires an explicit, off-by-default setting and a user-initiated action. - Retrieving a page needs a connection; working with an imported local source can continue offline. - Tenvar is not a web search service, a hosted email inbox or a cloud collaboration platform. ### Keep the work usable and portable. Listen to generated answers and other supported synthesized output with on-device Supertonic or Kokoro voices. Export a project as a portable, password-encrypted backup that can be restored into Tenvar. - Speech recognition and read-aloud use separate downloadable models. Language and voice choices depend on the read-aloud engine. - Continuous source-book read-aloud remains outside the 0.4.0 release. - A project backup is for restoration in Tenvar; document and report exports are the handoff for readers who do not use the app. ## Supported input formats ### Recordings Audio and video formats supported by the bundled media tools. Transcription, playback, speaker labels, marked moments and subtitles. Codec support varies by file. ### PDFs, text and books PDF, TXT, Markdown, EPUB. Read, extract text, search and cite. Scanned pages may need OCR. Protected or unusual book layouts can limit extraction. ### Office documents DOCX, RTF, PPTX, ODT, ODP; supported macro-enabled and template variants. Extract document or slide content. Office macros are not executed. Legacy DOC, PPT and other supported formats may need the optional document runtime. ### Spreadsheets XLSX, XLS, XLSB, XLSM, ODS, CSV, TSV and supported template variants. Read and extract tabular content for research. Tenvar does not replace a spreadsheet calculation or editing application. ### Images and scans PNG, JPEG, TIFF, BMP, WebP, GIF, SVG; HEIC and HEIF on macOS. Local OCR and supported visual tools. Recognition depends on image quality and the installed model. ### Email and comics EML, MSG, MBOX; CBZ and CBR. Saved emails and mail archives join the document workflow. Comic archives are treated as one source with pages. ### Existing notes Obsidian vaults; Apple Notes on macOS; supported Markdown exports. Import independent copies. No two-way synchronization or write-back to the original note library. ## Questions and limitations ### What is Tenvar, and who is it for? Tenvar is a private desktop workspace for research, transcription, notes and writing. It brings documents, recordings and imported notes into projects, uses on-device AI for transcription and cited questions, and turns selected evidence into notes, manuscripts and reports. It suits researchers, writers, students and people working with interviews, reference documents or client material who want to keep that work on their own computer. ### Is Tenvar an alternative to a cloud AI chat? Tenvar focuses on your own local sources: importing them, reading or transcribing them, asking cited questions, marking evidence and writing from it. It uses models installed on your machine rather than sending prompts and documents to a cloud AI provider. It is not a replacement for a general web-search assistant or a guarantee of the same capabilities as a hosted model. ### Does Tenvar work completely offline? Core reading, transcription, OCR, search, questions, writing and export work locally once the required models and components are installed. Model or document-runtime downloads, optional online web/email import, requested update checks, and paid-license activation or revalidation need a connection. Your source library is not uploaded for AI processing. ### Do I need an account or an AI API key? No account, credit card or cloud AI API key is needed to start using the Free plan. Models run on your own hardware. Buying a paid license is a separate purchase through Polar; the license key is then activated in the app. ### What does the Free plan include? Free includes one Tenvar project plus one imported library: choose either an Obsidian vault or Apple Notes. The imported-library allowance is shared, not one of each. Files within those allowances and the available tools are unlimited, with no time limit, transcription-minute quota, export watermark or feature paywall. Apple Notes import is Mac-only. ### What does the paid license unlock? The $59 USD early license removes the project and imported-library limits. It is a one-time purchase that includes version 1 and its updates; the planned version-1 price is $129. It is not a subscription. Activation and periodic validation check the license; being offline does not disable an already activated license. ### Do I need a powerful computer or an NVIDIA graphics card? A dedicated graphics card is not required. An 8 GB computer can use smaller models, and Tenvar checks memory and hardware when offering models. Larger speech, language and vision models need more RAM or VRAM and storage; some are unsuitable for smaller machines. Speed varies. Standard Windows and Linux builds support CPU and compatible Vulkan graphics; optional NVIDIA builds add CUDA. Apple Silicon uses Metal where supported. ### Is my work encrypted automatically? No. New projects start unencrypted. Choose Encrypt project and set a project password or PIN to protect that project. This secret is separate from your computer login, and Tenvar cannot recover it if forgotten. Original files outside Tenvar are not encrypted by importing them. App access locking and project encryption are different protections. ### Does Obsidian import keep my vault synchronized? No. You can browse a vault through Shelves and bring selected files into a project, or convert it into an editable Tenvar copy. The original vault is left alone. Later edits are not written back, and Obsidian plugin behavior or every formatting convention is not guaranteed to carry over. ### How does Apple Notes import work? On macOS, choose individual notes from Shelves or import an Apple Notes library. The app requests the system permission needed to read Notes and creates local copies. Rescan can bring in updates while preserving conflicting Tenvar edits. Text and available attachments have fidelity limits; locked or unreadable notes are skipped and problems are reported. It is not a full-fidelity Apple Notes backup or two-way sync. ### Can the people receiving my report use it without Tenvar? Yes. A self-contained HTML report can include selected playable audio and video evidence and can be opened in a browser offline without installing Tenvar. PDF is available for a printable handoff, but it does not preserve interactive media playback. ### Can I trust every AI answer or speaker label? No. Models can misunderstand speech, assign a speaker incorrectly or produce an inaccurate answer. Review the transcript and speaker assignments, and open citations to compare answers with the original. Language coverage, OCR quality and output quality depend on the model and source material. ### What are the current limits of version 0.4.0? Tenvar is early software and can have rough edges. There is no hosted collaboration or cloud library sync, no two-way Obsidian or Apple Notes sync, no live meeting captions, and no system-wide dictate-anywhere tool. Continuous source-book read-aloud and report-wide AI editing are not in this release. Hardware, import fidelity and document runtimes impose additional practical limits. ## Published installers - [Tenvar_0.4.0_aarch64.dmg](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_aarch64.dmg) — 115 MB - [Tenvar_0.4.0_amd64-nvidia.AppImage](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_amd64-nvidia.AppImage) — 1.32 GB - [Tenvar_0.4.0_amd64-nvidia.deb](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_amd64-nvidia.deb) — 1.24 GB - [Tenvar_0.4.0_amd64.AppImage](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_amd64.AppImage) — 275 MB - [Tenvar_0.4.0_amd64.deb](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_amd64.deb) — 175 MB - [Tenvar_0.4.0_x64-nvidia-setup.exe](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_x64-nvidia-setup.exe) — 791 MB - [Tenvar_0.4.0_x64-setup.exe](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_x64-setup.exe) — 141 MB - [Tenvar_0.4.0_x64_en-US.msi](https://github.com/jBlanca/tenvar.app/releases/download/v0.4.0/Tenvar_0.4.0_x64_en-US.msi) — 229 MB Mac: Apple Silicon, macOS 14.4+. Windows: x64, Windows 10 (2004+) or 11. Linux: x86-64, Ubuntu 24.04 / glibc 2.39+ baseline and compatible system dependencies. Mac installers are signed and notarized. Windows installers are currently unsigned. Models download separately.