Your recordings turned into notes, privately
Speakr transcribes your meetings and voice memos, then transforms them into organized and searchable notes thanks to AI. Installed and maintained by DINAO — your audios remain on our French servers.
What is Speakr?
Speakr is a personal and self-hosted web application for audio transcription and intelligent note-taking. It transforms your recordings — meetings, interviews, voice memos — into organized, summarized, and fully searchable notes, while running on your own infrastructure to preserve the confidentiality of your exchanges.
Speakr accepts various inputs: microphone, system audio, or browser tab, mixing both, or simple file import. Long sessions are streamed continuously to the server and survive a reload. On the speech recognition side, several engines are available (WhisperX GPU-accelerated, OpenAI Transcribe, Mistral Voxtral, VibeVoice), with speaker identification.
Designed for teams, Speakr offers multi-user management, groups with shared tags, and SSO/OIDC authentication (Keycloak, Azure AD, Google, Auth0…). Hosted and maintained by DINAO in France, it becomes a truly sovereign AI note-taking solution.
Host Speakr at DINAO
Resource tiers compatible with Speakr prerequisites (minimum 2 vCPU / 4 Go / 10 Go). Hosted in France, fully managed.
- 2 dedicated vCPU
- 4 Go RAM
- 40 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 4 dedicated vCPU
- 8 Go RAM
- 80 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 8 dedicated vCPU
- 16 Go RAM
- 160 GB NVMe
- Daily backups
- Managed & monitored by DINAO
1 hidden tier(s) (insufficient resources for this app) : Découverte
This application uses AI
The container hosts the application, not the AI engine (which requires dedicated GPUs) : inference is handled externally, with your own provider key. Prioritize a sovereign engine — Mistral AI, NumSpot, Scaleway, or OVHcloud AI Endpoints (France, GDPR) ; an international provider (OpenAI, Anthropic…) only if a specific capability requires it. Inference subscriptions are not included in hosting.
Technical details
You might be wondering…
Do my recordings go to an external service?
In local mode (WhisperX on your instance), no: everything stays in France on your server. If you choose an external transcription API, the audio is processed by that provider — to be avoided for sensitive content.
Does Speakr distinguish speakers?
Yes. Depending on the chosen engine, the transcription includes diarization, i.e., the identification of different speakers.
Can I record long meetings?
Yes. Long sessions are streamed continuously to the server and survive a page reload.
Can multiple people use Speakr?
Yes, with multi-user management, groups, shared tags, and SSO/OIDC authentication.
Are my notes searchable?
Yes. Transcriptions are indexed, summarized, and structured by AI to quickly find information.
