Many speech applications provide a worse user experience and task completion rate than they should, simply because they haven't been properly designed or tuned. Now would be a good time to get the return on investment you initially planned for.
The foundation of a great speech application is great dialogue design. Application requirements, dialogue specifications, persona definitions, voice talent coaching, we do it all.
No one is better qualified than Nu Echo to develop high-performance speech grammars for your project, irrespective of the type of grammars (SRGS, SLM) or speech recognition engine you use.
Whether you need to address speech recognition issues with a specific grammar or you need to improve the performance of a complete suite of applications involving hundreds of grammars, Nu Echo has the team, the tools and the processes to make it happen.
You need to migrate your speech application to a new recognition engine? No problem! With our specialized tools and proven processes, Nu Echo can handle conversion projects with hundreds of grammars in record time.
Know more about your customers and get insights about applications that are currently deployed. Put system optimization on the fast track.
Nu Echo offers the full gamut of speech consulting services. Our areas of expertise include speech recognition, speech synthesis, natural language processing, and voice biometrics.
Optimize user experience by validating system design and dialogue with real end users. Nu Echo's expertise in usability testing and user-centered dialogue design will ensure the best possible user experience, task completion rate and customer satisfaction.
Take advantage of Nu Echo's deep expertise in multilingual text-to-speech development and tuning to achieve optimal performance with your speech synthesis engine. Make your application sound as good as it can!
Transforming raw data into usable data sets, from phonetic and orthographic transcription to semantic tagging.
Atelier is Nu Echo's proprietary speech application tuning environment, used by our speech scientists to enhance the performance of our customers' applications. It provides a complete suite of data analysis and diagnostic features that enable us to quickly and efficiently target problem areas, and to find and test solutions in a reliable iterative process. With complete multi-engine support in a single environment, Atelier can tune any application.
Tuning a speech application requires the ability to perform off-line speech recognition experiments iteratively, with the grammars – static or dynamic – used by the application, and to analyze the results in order to improve speech recognition accuracy, set confidence thresholds, test post-processing algorithms, etc.
As the developer of the widely used NuGram IDE Basic Edition, Nu Echo's speech grammar development capabilities are respected throughout the industry. In order to most effectively deliver on grammar development or conversion projects, Nu Echo's speech science team relies on our proprietary NuGram IDE Pro Edition, which includes advanced features specifically for grammar conversion and testing.