Speech to speech
Palaver goes from audio to meaning to audio, so it can serve languages that are rarely or never written.
Palaver Voice
Palaver Voice.
Real-time and on-demand speech translation, transcription and generation, built first for the institutions that serve people and, later, for everyone.
In development Concept stage. Details may change as we build.
Concept preview
What talking to Palaver could feel like.
Tap the mic to step through a sample exchange: listening, thinking, then speaking. Switch the caption language, pick a spoken language or open the transcript.
Tap the mic to start
- The conversation transcript will appear here, with the original language and the translation side by side.
Illustrative sample. No audio is recorded or played, and nothing you do here is sent anywhere. Speech in the selected language is represented by its translation.
What it is
Speak your language. Be understood.
Palaver Voice is the conversational face of the language layer: speech translation, transcription and generation in the languages people actually speak.
Today, health ministries, NGOs and governments produce materials mainly in colonial languages and then try to distribute them. Uptake is limited by comprehension. Palaver Voice is being built to close that gap, so a message can reach people in their own language and the answer can come back the same way.
See the conceptHow it's designed
Voice in. Voice out.
Live captions
Transcription and translation into English or French, for the clinic, the office or the record.
One message, many languages
A content engine that turns a single source message into localized audio and video across hundreds of languages.
Coverage you can report
A coverage dashboard built for the reporting that ministries and NGOs already do.
Careful by design
In high-stakes moments like health or money, Palaver asks for confirmation rather than acting when its confidence is low.
Honest about limits
Transparency about what the system can and can't do is part of the product, not an afterthought.
Who it's for
Built for the people who must be reached.
- Health ministries and NGOsReach the people campaigns and clinics already intend to serve, in the languages they speak.
- GovernmentsRender a single message with the right register across many linguistic communities.
- Community health workersHave a conversation in the patient's language, with a written record in English or French.
- Everyone, laterConsumer voice features are planned once quality and coverage are proven.
What's coming
The road ahead.
Where we're headed. These are plans, not shipped features.
- First
A voice-first health pilot
A pilot voice-first health application with a ministry or NGO partner in Cameroon, measured by how many more people it reaches.
- Next
The content engine
One source message turned into localized audio and video across hundreds of languages, with a coverage dashboard for reporting.
- Then
Conversation in real time
Streaming translation with sub-second lag for natural conversation, plus robustness across dialects.
- Later
Voice for everyone
Consumer voice features and real-time cross-border voice translation for trade.
Also in the Palaver family
Explore the other products
In development
Palaver AI
Core models and APIs for language understanding, generation and agentic tasks across African languages.
Explore Getting startedPalaver Labs
Research, data operations, community partnerships and long-term model development.
Explore On the horizonWhat's next
Once the language layer exists, new products become applications of it: healthcare delivery, government communication, voice-first financial services and ambient devices.
Explore