VOCAPIA is an R&D company and software publisher developing leading-edge speech processing technologies for most European languages, as well as Arabic, Mandarin, Russian, Ukrainian.
VOCAPIA provides valuable high performance tools for the exploitation of audio data. The Voxsigma software targets two types of audio: broadcast speech and conversational telephonic speech.
Main applications: lawful interception, intelligence analysis for OSINT and COMINT, audio mining, topic spotting, media monitoring.
Filtering information in audio data is critical for national security in order to ensure the protection of citizens and as such is a strategic need for many governmental agencies. Vocapia develops leading edge technologies to automatically transform large volumes of unstructured audio data to meaningful textual information. The VoxSigma software suite is available for licensing and as a web service.
VoxSigma first indentifies the audio segments containing speech, then it recognizes the language being spoken if it not specified a priori, indentifies the speakers (localizing all segments for a given speaker), and finally converts the speech segments to text including punctuation, case, numeric and the associated time-codes.
The XML output file can be directly indexed by a search engine for audio and audiovisual data mining applications, or alternatively can be converted into plain text.
D12 - Software - Cloud solutions
28, rue Jean Rostand - Parc Orsay Université
91400 Orsay
France
+33 (0)1 84 16 91 84
+33 (0)1 60 19 54 94
VOCAPIA est une entreprise de R&D et un éditeur de logiciels spécialisé dans le développement de technologies de traitement automatique de la parole pour la plupart des langues européennes ainsi que l’arabe, le mandarin, le russe et l’ukrainien.
La suite logicielle VoxSigma met en oeuvre des technologies de pointe pour segmenter les documents audio, identifier la langue parlée et transcrire la parole pour transformer des données audio brutes en documents structurés de type XML facilitant ainsi la recherche d’information.
Principale applications: Analyse de données conversationnelles téléphoniques, pige audio visuelle, fouille de données audio, indexation multilingue de données audio.
Fondée en 2000, VOCAPIA a depuis l’origine un partenariat privilégié avec le LIMSI, laboratoire du CNRS de renommée mondiale dans le domaine du traitement de la parole. Grace à ce partenariat VOCAPIA peut mettre rapidement sur le marché des produits faisant appel aux dernières avancées de la recherche et participer à des projets de recherche.
Filtering information in audio data is critical for national security in order to ensure the protection of citizens and as such is a strategic need for many governmental agencies. Vocapia develops leading edge technologies to automatically transform large volumes of unstructured audio data to meaningful textual information. The VoxSigma software suite is available for licensing and as a web service.
VoxSigma first indentifies the audio segments containing speech, then it recognizes the language being spoken if it not specified a priori, indentifies the speakers (localizing all segments for a given speaker), and finally converts the speech segments to text including punctuation, case, numeric and the associated time-codes.
The XML output file can be directly indexed by a search engine for audio and audiovisual data mining applications, or alternatively can be converted into plain text.
2000: Année de création
300 000: Nombre d'heures de parole transcrites en 2015 via notre web service
30: Nombre de pays utilisant nos logiciels
24: Nombre de langues traitées
3.1: Chiffre d'affaire 2015 en M€
45: Pourcentage export du chiffre d'affaire
D12 - Logiciels et Solutions "cloud"
28, rue Jean Rostand Parc Orsay Université
91400 Orsay
France
+33 (0)1 84 17 01 14
+33 (0)1 60 19 54 94