Ensure that your content reaches the widest audience, even on those breaking news. AUDIMUS.MEDIA adoption is the cost–effective solution for your live subtitling needs.
AUDIMUS.MEDIA produces subtitles automatically and in real time on live contents.
300+ Happy Customers Worldwide
Earl Flanigan
Daniel Fernandes
We help companies to make their live content accessible
Broadcast & Media
Meet the regulations for live subtitling. AUDIMUS.MEDIA complies with all accessibility regulations for TV and internet broadcasts, in terms of accuracy and promptitude.
Corporate
Make your content accessible to all your employees. Connect with everyone by live streaming your event, webinar or conference with subtitles.
Education & Culture
Assure equal access to knowledge. Universities, schools and e-learning platforms can deliver content with subtitles in real time for their lectures, meetings, conferences.
How live subtitling works
Straightforward integration with your production workflow
CAPTURING
AUDIMUS.MEDIA captures feeds from any pre-existing or produced audio or video' sources (digital or analog).
PROCESSING
Through automatic speech recognition, AUDIMUS.MEDIA produces subtitles automatically with a very small latency time, identifying all pivots and different speakers.
DELIVERING
The system will send the subtitles to a subtitling/teletext encoder, in less than 3-4 seconds’ word by word. The output text is standardised, formatted and optimized for a wide range of encoders.
In depth view
We provide different language models with specific themes each (e.g. news, sports). These are trained with large amounts of audio and text, where the total volume of vocabulary exceeds 200,000 words. Even more, it is possible to adjust the phonetic transcriptions to adapt models to local language.
Supported languages:
English (American, Canadian and British), European Portuguese, African Portuguese, Brazilian Portuguese, Spanish (Castilian & Latin America), French, German, Italian, Swiss and Basque.
Years of R&D led VoiceInteraction to create its own speech-to-text innovative technology. Highlights:
- Standard language models >200,000 words dictionaries while maintaining real-time processing
- Maximum accuracy with the adaptation to the most frequent voices, daily update of the vocabulary context.
AUDIMUS.MEDIA allows the addition of metadata automatically to the content to be broadcast:
- Speaker ID: the indication of who is the speaker ( [ ] )
- Speaker Turn: the indication of speaker changing ( >> )
- Language ID: the indication of speaker language
- Language Turn: the indication of when the language changes and automatically changing the language model in use
Do you need to add captions in your broadcast content? We deliver you a transcript with time stamp so that you can focus on what matters: make use of your contents for other ends, such as internet publishing.
AUDIMUS.MEDIA exports a subtitle file, which will be automatically synchronized with your audio. There's no need to review your transcript. The system will eliminate the latency time.
Hardware Requirements
- Windows 7/8/10/WS2012/WS2016
- Intel Core i7 or Intel Xeon with at least 4 cores (physical), CPU Freq >2.5GHz
- ≥ 8GB of Ram
- ≥ 5GB of free disk space
- SDI Declink card (to use directly SDI - preferably) or Sound card for balanced audio
System Administration
- Web interface