SpeakScriber
SpeakScriber
The fastest way to turn a spoken conversation into text anyone can read, and text into speech anyone can understand, live, across more than 70 languages, in your browser today, with a mobile app close behind.
Every day, professionals, travellers, clinicians and creators run into the same wall. They need to understand someone speaking a different language right now, in the moment, without booking a human interpreter or wrestling with an app that only handles typed text.
SpeakScriber listens as you talk, transcribes every sentence the moment you finish it, and translates that sentence into your listener's language, out of more than 70 supported, before the conversation moves on. Today that runs entirely in your browser, nothing to install, nothing to configure, with the same experience coming to a native mobile app next.
Talk normally in your own language. SpeakScriber captures audio straight from your microphone in the browser tab.
Nova-3 converts speech to text in real time, holding the same accuracy across accents and dialects, and commits each sentence the moment you pause.
Microsoft Translator or DeepL renders that sentence in their language instantly, with the option to read it aloud.
Works two ways: live, in the middle of a real conversation, or after the fact, from a recording you already have. The next slide shows exactly what that looks like.
Every sentence appears on screen the instant you stop speaking, and it stays accurate whatever your accent. Deepgram Nova-3 listens continuously and commits each phrase to text in a fraction of a second, with no compromise for regional or non-native speech.
The moment a sentence is transcribed, it is translated into your listener's language automatically, across more than 70 languages, powered by Microsoft Translator with DeepL for the pairs that need extra precision.
Already have a recording? Drop in an interview, lecture or consultation as MP3, WAV, M4A, FLAC, OGG or WebM, and get back a full transcript with translation and timestamps in minutes.
Every translated sentence can be read aloud on demand, and every saved session can be replayed later, complete with a karaoke style highlight that tracks the words as they play.
Every conversation and every upload is saved automatically. Search past sessions and export any transcript as a PDF or plain text file whenever you need a record.
Share a workspace with up to five seats, see usage per member, and give a whole team the same translation quality from one shared account.
Different jobs, the same problem: understanding someone in real time, in a language that isn't your own.
Executives, sales reps and consultants crossing language borders every week, who need real help in meetings and negotiations, not a phrasebook.
Adults actively learning a European language who want to see what they just said transcribed, then compare it instantly against the correct translation.
Doctors, nurses and lawyers working with non-native-speaking clients, where accuracy and privacy cannot be compromised. This segment justifies the higher-tier plan.
YouTubers, podcasters and live streamers who need real-time captions and translation for an international audience, expensed against channel income.
People who record first and transcribe later. An hour pack replaces paying a few euros a minute for a human transcription service, per interview.
Live Conversation Mode turns a single phone into a real time interpreter for two people. Each side speaks their own language into their half of the screen, and hears the other person's reply spoken aloud automatically, in their own language, with no menus and no switching.
Two markets are converging on SpeakScriber, and the fastest growing slice of both is exactly where we compete.
Hour packs are built for the traveller, the conference attendee, or anyone who needs SpeakScriber for one important event rather than every single day.
| Pack | Hours | Price | Effective rate | Bonus |
|---|---|---|---|---|
| Starter | 1 hour | €1.99 | €1.99 / hour | None |
| Standard | 5 hours | €7.99 | €1.60 / hour | None |
| Plus | 10 hours | €13.99 | €1.40 / hour | +2 hours free |
| Power | 25 hours | €29.99 | €1.20 / hour | +5 hours free |
Pro and Team plans include a monthly hour allowance at a lower effective rate than any pack, plus a discounted top up rate if the month runs out early.
| Plan | Included hours | Daily cap | Price | Top up rate |
|---|---|---|---|---|
| Free trial | 1 hour total, once | None | Free | Not applicable |
| Pro | 30 hours / month | 4 hours / day | €14.99/mo or €149/yr | €0.90 / hour |
| Team, 5 seats | 20 hours / seat / month | 6 hours / day / seat | €49 / month | €0.80 / hour |
| Enterprise | Custom hour pool | Custom | Custom pricing | Custom |
Ten development phases are already shipped, tested and live. What is missing is not engineering. It is customers, and that is exactly what this prize helps fund.
A design token system and a premium visual pass across the whole product. Phase 11.
Revenue, usage and API cost dashboards. The operating system for running the business at scale. Phase 12.
React Native and Expo, iOS first. Live Transcribe, Conversation Mode and the offline recorder. Phase 13.
Same codebase, four to six weeks after iOS ships. Enterprise outreach begins once Year 2 traction is validated.
Final user acceptance testing across web and mobile, then the organic marketing push from the business plan: Product Hunt, Reddit, YouTube, and the affiliate programme. Phase 14.
Designed and built the entire product from the ground up: architecture, database, authentication, billing, the real time speech pipeline, and the mobile roadmap that comes next.
Owns sales and marketing end to end: go to market execution, channel strategy, and turning a finished product into paying customers.
Prize funding and mentorship to fund the mobile build and our first paid marketing push, turning a finished product into the 4,000 sign ups and 120 Pro subscribers already modelled for month 12.
Reference material for written review and questions and answers. Not part of the live pitch.
| Tool | Pricing | Strength | Weakness |
|---|---|---|---|
| Google Translate | Free | Ubiquitous | No live transcription, no history |
| DeepL Pro | €8.99 to €57/mo | Best translation quality | No live speech, no transcription |
| Otter.ai | $10 to $20/mo | Great transcription and upload | English only, no real time translation |
| Sonix | $10/hr or $22/mo | Good file transcription | No live mode, translation costs extra |
| Trint | $48+/mo | File transcription for journalists | Expensive, no live mode |
| Wordly AI | $99+/mo | Enterprise live events | Far too expensive for individuals |
| KUDO AI | $249+/mo | Professional interpretation | Enterprise only |
| SpeakScriber | €0 to €15/mo | Live and upload, transcription and translation across 70+ languages | New entrant, no mobile app yet |
| Year | Sign-ups | Pro subs | Team | MRR | ARR |
|---|---|---|---|---|---|
| Year 1 (month 12) | 4,000 | 120 | None yet | ~€317 | ~€16,000 total revenue |
| Year 2 | 15,000 | 450 | 25 workspaces | €7,970 | €95,600 |
| Year 3 | 40,000 | 1,200 | 80 workspaces, 5 Enterprise | €22,900 | €274,800 |
Break-even is reached at around 9 Pro subscribers. Source: docs/BUSINESS_PLAN.md, section 6.
At the month 12 scale modelled here, each Pro subscriber costs about €1.83 a month to serve, Deepgram usage plus a share of infrastructure, against €14.99 in revenue. That is roughly 88% gross margin per subscriber, and it recovers the entire cost of acquiring that subscriber within the first month.
Costs include Deepgram Nova-3, Microsoft Translator (2 million free characters a month), Google Cloud TTS, VPS hosting and PayPal fees of 3.49% plus a fixed fee. Source: docs/BUSINESS_PLAN.md, sections 6 and 8.
| Phase | Focus | Status |
|---|---|---|
| 1. TypeScript migration | Full stack TS conversion | Shipped |
| 3. CI/CD | GitHub Actions pipelines | Shipped |
| 4. Database | PostgreSQL and Prisma | Shipped |
| 5. Auth | JWT, Google OAuth, email flows | Shipped |
| 6. History and export | Session history, PDF/TXT, TTS | Shipped |
| 7. i18n | 15 language UI | Shipped |
| 8. Subscriptions | PayPal billing, hour packs, credits | Shipped |
| 9. Team plan | Workspaces, seats, admin dashboard | Shipped |
| 10. Audio upload | GCS and Deepgram pre-recorded pipeline | Shipped |
| 11. UI polish | Design tokens, premium visual pass | In progress |
| 12. Admin backoffice | Revenue, usage and cost dashboards | Planned |
| 13. Mobile app | React Native and Expo, Conversation Mode | Planned, about 9 weeks to iOS |
| 14. UAT and launch | Final QA, then the organic marketing push to first customers | Planned |
SpeakScriber, live speech transcription and translation. speakscriber.app