SermoVox
SermoVox is a real-time multilingual speech translation system for live events. It converts a speaker's speech into text and delivers translations to the audience with minimal delay.
SermoVox Local is available as a locally installed software product with a one-time licence fee.
What is it?
SermoVox is a locally operated speech recognition and translation system developed by FaktumAI. It is designed for situations where the same speech needs to be delivered to an audience in multiple languages with as little delay as possible.
The current production direction focuses on recognising Finnish speech and delivering translations in English, Ukrainian and Russian. The system creates one continuous pipeline from microphone input to speech recognition, translation and audience-facing subtitles.
Price and licence
€599 + VAT
Permanent one-time licence
SermoVox Local is purchased with a one-time payment. The customer receives a permanent licence to the delivered local SermoVox software version on one agreed workstation. The delivered version does not require an ongoing monthly or annual subscription.
What is included?
- permanent licence for SermoVox Local on one agreed workstation
- software installation and initial deployment
- deployment of the required speech recognition and translation models
- testing of the microphone or headset and display or projector setup
- deployment of the operator and audience views
- basic user guidance
- standard user support on weekdays and by separate agreement
On-site deployments are currently available in the South Ostrobothnia region of Finland.
What problem does it solve?
Professional interpretation for multilingual events can be expensive or difficult to arrange. Consumer translation applications, on the other hand, are not designed for continuous speech, operator-controlled workflows and real-time subtitles shown on a large display.
SermoVox is intended to provide a lightweight, manageable and primarily local solution for this use case.
How it works
Microphone → speech recognition → translation → real-time data transfer → operator and audience views
The speaker's audio is processed locally. Speech is recognised, translated into the selected languages and delivered to a browser-based interface. The operator controls the system through a dedicated view, while the audience sees the translations through a separate projector or display view.
Who is it for?
- churches and multilingual communities
- events and seminars
- associations
- organisations that need multilingual communication
- municipalities and other public-sector organisations that need a locally controlled translation solution
Hardware
The computer, microphone or headset and any required display or projector hardware are not included in the €599 software price.
SermoVox can be installed on a suitable customer-owned Windows computer. FaktumAI can define the required hardware configuration and assist with selecting suitable equipment before deployment.
Hardware requirements depend on the languages used, the speech recognition model and the required level of performance. The requirements are reviewed before delivery.
Technical implementation
The current SermoVox implementation uses GPU-accelerated speech recognition, local translation models, a FastAPI backend and WebSocket-based real-time data transfer. The user interface consists of a dedicated operator view and a separate audience-facing projector view.
The system is designed around local processing, low latency, operational reliability and reduced dependence on external cloud services during production use.
Permanent licence
SermoVox Local is delivered with a permanent one-time licence. The customer's right to use the delivered software version does not expire if the customer chooses not to purchase future versions or additional services.
Future major product versions, new paid features, additional languages or separate cloud services may be priced separately.
Updates
Corrections and compatibility updates may be provided as the product develops. The update model is still being refined, but the permanent licence does not depend on a recurring subscription.
Major future product versions or entirely new services are not automatically included in the original one-time licence.
Support
Standard SermoVox user support is normally available on weekdays and at other times by separate agreement.
Standard support covers guidance related to normal software use. More extensive on-site work, hardware changes or other separately agreed services may be priced separately.
Current status
SermoVox is available. The first customer deployment has been agreed and preparation is underway for installation in the customer's environment. Current development focuses on deployment readiness, speech recognition latency, audio-path reliability and verification of a production-ready hardware configuration.
Next step
The next major step is to install SermoVox in the first customer's environment and test it in a real event. Experience from the deployment will be used to refine the delivery process, user experience and further product development.
Solutions for organisations and larger deployments
The €599 + VAT SermoVox Local package is intended for deployment on one agreed workstation. For organisations requiring multiple workstations, multiple locations or a broader implementation, the scope and pricing are planned separately according to the customer's requirements.
SermoVox can be deployed in the customer's own local IT environment. In such a deployment, speech recognition and translation can be processed on customer-controlled hardware without sending speech data to an external cloud service. This can be important for organisations with stricter requirements concerning privacy, security or control of data.
- multiple SermoVox workstations
- multiple deployment locations
- centrally coordinated deployment
- customer-owned hardware
- customer-specific hardware configuration
- local or customer-controlled server environment
- customer-specific support arrangements
- customer-specific language and operating-environment requirements
- potential future cloud or hybrid deployment
Data under customer control
A key advantage of a local SermoVox deployment is the ability to process speech and translations within the customer's own environment. FaktumAI can design the deployment model together with the customer according to its IT and information-security requirements.
Need multiple devices or an organisation-specific deployment?
Larger SermoVox deployments are planned according to the customer's use case, number of devices, languages and security requirements.
SermoVox Cloud — future option
Alongside the locally installed SermoVox Local product, FaktumAI is investigating a cloud-based service model. The long-term goal is to provide an alternative for customers that want to use SermoVox as a browser-based or cloud service without maintaining their own local AI environment.
SermoVox Cloud has not been released and currently has no confirmed pricing or launch schedule.
Need real-time multilingual translation?
SermoVox Local is available as a locally installed solution with a one-time licence fee. We can review the use case, required languages, hardware and suitable deployment model together. SermoVox Local €599 + VAT
marko@Faktum-AI.com