Automatic Live Subtitling; Spoken Dialogue; Voice Recognition and Text-to-Speech
Our organisation type is:: SME
We are active in the following Technologies: Information technologies, data storage and processing (back end), Interfaces, software (front end), Communication technologies / Tele-systems, Learning & Training, Accessibility
Applications: Home care, Information and learning, Knowledge Transfer, Safety, security, privacy, Skills enhancement, Social interaction
VOICEINTERACTION Speech Processing Technologies, S.A., is a technology-based firm operating at the global market through the development of applications based on its speech synthesis and recognition engines, DIXI and AUDIMUS, respectively. Due to the AUDIMUS capability to operate, not only in real-time, but also in online mode, VoiceInteraction created a fully automated broadcast news transcription system, supporting European and Brazilian Portuguese, Castilian Spanish, Basque, French, Italian and German and all Swiss varieties of the latter three languages. Also, a Multimedia Monitoring Solution was built on AUDIMUS online transcription capabilities in order that the "stories" (occurring in the TV channels or radio stations) matching the clients' profiles can be sent them as soon as they are broadcast. In speech synthesis, VoiceInteraction has been involved in automatic methods for creating customized synthetic voices using either professional or non-professional voice talents (i.e. coping with
speaker-specific post-lexical phonological processes by automatic means in order to increase the size of the concatenation units). VoiceInteraction was established in 2008 as a spin-off from the Spoken Language Systems Lab (L2F) of INESC-ID Lisbon. The strong connection between these two entities has proven to be one of the company's success factors, since it makes it possible for both teams to share expertise in multiple topics of common interest.