On this website we use cookies, even from third parties, to improve your visit. 

Hi in PerVoice languages

 

 

Trento, 01 March 2018 - Time for a monthly update of PerVoice language models!


In the past few months we developed Swiss French, Swiss German and updated the Swiss Italian models for broadcast monitoring.
These models are essential to cover the entire Switzerland, with all its dialects and accents. This will be a huge step to allow PerVoice's partners to expand their Swiss business.
We also updated our main English models, both UK and US ones. These updates improve the accuracy of transcriptions and slightly increase the performances using less computing power than the previous one.

Moreover, we updated Austrian German model, Dutch model and Greek model: all developed within Broadcast news domain. Last but not least we developed Colombian Spanish model for call centers, already successfully used by one of our main partners. For further information fill out the form in the "Contact Us" page (http://www.pervoice.com/en/contact-us).

Hi in PerVoice languages

 

 

Trento, 2 January 2018 - Based on the experience gained in the development of the European French model, PerVoice SpA can proudly announce the availability of the new French Canadian language model. This adds to the other 29 languages supported by the PerVoice Speech-to-Text engine, which is successfully used in many voice recognition applications such as Broadcast Monitoring, Real-time Transcription and Interactive Voice Response (IVR).


The French Canadian language model, included in the 2017 roadmap for PerVoice activities, is intended to cover a huge part of the American continent and is well performing on the pronunciation of that specific area. Successfully build by the Language Technology team, the model is well suited to transcription in Broadcast Monitoring scenario is available for the use with AudiomaBox and Audioma 2. For more information you can reach us by filling out the form in the "Contact Us" page (http://www.pervoice.com/en/contact-us/).

 

Trento, 13th July 2017 - PerVoice SpA is pleased to announce the development of the new language model for voice recognition of Urdu is completed. This adds to the other 25 languages supported by the PerVoice Speech-to-Text engine, which is successfully used in many voice recognition applications.

The Urdu language model, included in the 2017 roadmap for PerVoice activities, will be an important tool for consolidating voice recognition technology with Middle Eastern languages. Successfully build by the Research & Development team, the model is well suited to transcription in Broadcast Monitoring scenario is available for the use with AudiomaBox. For more information you can reach us by filling out the form in the "Contact Us" page (http://www.pervoice.com/en/contact-us)

PerVoice - Viale Verona 190/1 38123 Trento Tel. +39 0461383960 - Email:info@pervoice.it - Part. IVA 02019100227          ©2014 PerVoice SpA

PerVoice is a company of