Record voice notes
Tap-to-record directly in the app — meetings, dictation, ideation.
Vocally captures content across nine input channels — voice, meetings, YouTube, PDFs, presentations, documents, web links, in-app notes, and hardware feeds — organizes it with hierarchical multi-tags, and lets you query everything through Magic Chat. One app. Every format. One memory.
Most AI note-takers stop at meetings. Vocally captures across nine input channels — voice, meetings, audio uploads, YouTube, PDFs, presentations, documents, web links, and in-app text — plus three hardware companions. Everything gets organized through hierarchical multi-tags. Then Magic Chat lets you query a single transcript, a group of them, or everything tagged under a label — all as one conversational scope.
Other AI productivity apps optimize for one input channel — usually meetings — and the rest of your day's information lives outside the app. Vocally captures the full surface area of what you encounter and makes it queryable.
Tap-to-record directly in the app — meetings, dictation, ideation.
In-room recording or virtual meeting capture — full transcript automatically.
Existing call recordings, podcasts, interviews — any audio file.
Paste any YouTube URL — Vocally pulls audio + transcript automatically.
Research papers, contracts, textbooks — uploaded and indexed.
Slides, decks, lecture material — text and structure extracted.
Word docs, text files, structured documents.
Articles, blog posts, web research — fetched and indexed.
Type or paste notes directly in the Vocally notes section.
Most AI chat assistants work on a single transcript. Magic Chat works at three scopes — pick the one that fits the question you're asking.
Chat with a single recording, document, or video. Ask for a summary, surface key quotes, generate action items — anything scoped to one source.
E.g., one meeting · one PDF · one YouTube video
Select several items and chat across them. Useful when content spans multiple sources but doesn't yet have a shared tag.
E.g., three interviews · two articles · one document
Chat with everything tagged under a label. The tag pulls every transcript, video, PDF, document, and note tagged with it into one conversational scope.
E.g., "Chemistry Sem 1" · "Q3 client meetings" · "Election interviews"
Student records 60 chemistry lectures across one semester. Tags each one "Chemistry Sem 1." Also uploads two textbook PDFs, six YouTube revision videos, and their personal handwritten note photos — all tagged the same. Magic Chat at the tag level becomes a personal tutor that has read everything. "Explain organic chemistry chirality, citing the lecture and the textbook." It cites both.
Vocally's tagging model is a graph, not a tree. Every captured item can carry multiple tags. Every tag can group items across formats — audio, video, PDF, document, web link, text. Magic Chat operates on the tag's full content, not just the format you started with.
Vocally pairs with hardware built by Brandworks Technologies — our co-development partner. The device records, the audio streams to Vocally servers, and the user sees notes + Magic Chat threads on the paired app. Three device classes; one Vocally pipeline.
Discreet device that records meetings, calls, and conversations. Audio streams to Vocally server; the user sees notes and Magic Chat threads on the paired Vocally app on mobile.
Wearable form factor with always-on capture. Audio + context flows to Vocally for hands-free capture during meetings, lectures, field work.
Desktop / room speaker for ambient capture in offices, conference rooms, study spaces. Same Vocally app pipeline.
Each persona uses Vocally today; persona-specific intelligence features are on the roadmap. The capture-and-query foundation is the same; the value layers are vertical.
Tag lectures by semester and subject — Chemistry Sem 1, Physics Sem 2, etc. Magic Chat queries the entire semester as one knowledge base.
Mock-test generation from tagged study material.
Tag interviews by story or beat. Magic Chat across all sources tagged under a story to find quotes, themes, contradictions.
Article draft generation from tagged source material.
Tag client meetings by case or engagement. Magic Chat extracts arguments, commitments, deadlines across the engagement.
Auto-generated action items and meeting summaries per engagement.
Tag speeches and engagements. Magic Chat for theme tracking, message consistency, audience response patterns.
Speech analytics and engagement insights.
Vocally is the consumer face of the SandLogic full-stack. The same engine that powers enterprise voice agents and speech analytics runs the captures, transcription, and Magic Chat — with the option to deploy Vocally on-prem for enterprises that need data sovereignty.
Speech-to-text. −51% WER vs Whisper-large-v3. Powers every voice capture.
Sovereign small language models. Powers Magic Chat conversational layer.
Text-to-speech for any spoken-output mode in future Vocally features.
Runs the models. Same engine that drives +73% throughput vs vLLM in enterprise.
For enterprises (law firms, accounting firms, hospitals, regulated industries) where individual-user data cannot leave the perimeter — Vocally is deployable on-prem. The same app, the same Magic Chat, the same Brandworks hardware integration, but the captures and the reasoning happen inside your data center. Same architecture; different deployment.
App Store and Play Store launch is on the roadmap. In the meantime, beta access is invitation-based across web, iOS, and Android. Two paths:
Tell us your use case. Beta seats are being released in tranches.
Request individual beta →On-prem Vocally deployment for organizations where individual-user data cannot leave the perimeter.
Talk to sales →Brandworks AI note-taker hardware is also shipping to beta customers — the device comes paired with Vocally access. Reach out at info@sandlogic.com to learn more.