Bilingual Korean+English AI voice memo assistant with semantic search
Translation pending
ProductivityTypeScript · CSSGitHub @thecyprinceOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
GitHub star growth between the latest completed 8-hour observations
DeepSeek Harness Remote can securely connect a phone or another computer to the computer where you run DeepSeek Harness, so you can continue your work anytime and anywhere, check progress anytime, and continue working.
AI translation
Open-source, local-first task manager that works offline, syncs every device, runs on a Raspberry Pi, and requires no subscription.
AI translation
Open-source, self-hosted webmail built on JMAP. Email, calendar, contacts, and files in one interface for Stalwart Mail Server.
Open Source Office suite for users used to MS Office - multi-platform, tuned for performance
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
The maker's introduction consists of the line above. Read the repository README for more details. Repository README ↗
Loading the full README.
Read the full README ↗Bilingual Korean+English AI voice memo assistant. Record a memo in Korean, English, or both — get a transcript with per-segment language tags, an LLM summary, extracted action items you can add to Google Calendar, and a semantic search bar to ask questions across all your past memos.
Live demo: https://voicebridge-one.vercel.app
ko+en with code-switching enabled)[KO] or [EN] with one-tap translation| Layer | Choice |
|---|---|
| Frontend | Next.js 16 (App Router) + TypeScript + Tailwind + shadcn/ui |
| STT | Gladia API (pre-recorded, diarization + code-switching) — migrated from OpenAI Whisper API, which has no diarization |
| LLM | Claude Sonnet 4.6 (Anthropic SDK, prompt caching) |
| Embeddings | OpenAI text-embedding-3-small |
| Vector store | Upstash Vector |
| Database | Supabase Postgres |
| Audio storage | Vercel Blob |
| Auth | Clerk |
| Calendar | Google Calendar API v3 |
| Deploy | Vercel |
git clone https://github.com/thecyprince/voicebridge
cd voicebridge
npm install
cp .env.example .env.local
# Fill in all values — see .env.example for instructions
You'll need accounts for: Gladia, OpenAI, Anthropic, Upstash, Supabase, Vercel, Clerk, and Google Cloud (Calendar API).
Run supabase/schema.sql in the Supabase SQL editor.
npm run dev
# Open http://localhost:3000
# Populate evals/data/ with audio files + ground truth first (see evals/*.ts for format)
npm run evals
Note:
evals/wer.tsstill measures OpenAIwhisper-1, not the production Gladia STT — the WER targets below are from the Whisper era.
| Eval | Target |
|---|---|
| STT WER (English) | < 10% |
| STT WER (Korean) | < 15% |
| STT WER (code-switched) | < 20% |
| Summary faithfulness | ≥ 4.5/5 |
| Summary relevance | ≥ 4.0/5 |
| Action item precision | ≥ 0.90 |
| Action item recall | ≥ 0.70 |
| Retrieval P@3 | ≥ 0.70 |
| Retrieval MRR | ≥ 0.60 |
vercel --prod
# Set all env vars in the Vercel dashboard