Customer Service is Customization

Commercial voice recognition is a new exciting technology that has you ask a lot of questions. We will break down what voice does in terms that you understand regardless whether you are a newbie to voice technology or whether you have in-house speech experts. Also, customization and support is an integral part of every deal we make because we are aware that just throwing this new technology at you isn’t going to get you to your goals.

Voice Solution Customization

Your product is currently controlled by a visual/haptic interface (touch screen, buttons, keyboard etc), and you want to substitute this interface or add voice control. We collect the requirements (the functions voice should control), suggest optimizations and then demo a prototype that shows you how voice would improve the user experience of your product. If the idea works for you, we will then go ahead and customize SpeechFoundry (our voice software) to produce a production-level voice user interface.

Customizable Language and Semantic Models

Every component of our system can be customized to meet the specific needs of your project. This is a crucial advantage when it comes to building “language models” for your project. We can and do build “language models” and “semantic models” customized for each project – this is the only way to ensure constant and robust user experience. Please compare to the “free” voice APIs that some competitors offer: those APIs do not allow you to build specific language models, but you will end up using the generic “cover all” language models with low usability for your specific domain.


You can create, customize and test voice solutions yourself in our intuitive, browser-based development environment for SpeechFoundry, for free ! – just get yourself an account with our customer service. You don’t need to install any software, just login to our website, and after getting some fundamental settings done with the help of our team, start playing around by building simple voice commands yourself that you can test immediately with your own voice with a standard PC microphone – it is fun! If you want to use the voice solution that you and your engineers created on our website, we will be happy to help you optimizing the experience and then port it to your specific embedded or customized environments.

Transcription Service and Voice Search

Simply upload your audio or video files and wait to get an email with the transcription. Log in to search the transcription and jump to relevant parts of the audio. Or have the transcription solution run on a local device with even faster access to results. You can have us add characteristic vocabulary for your domain (i.e. medical terms, customer center dialog phrases) to our dictation language models to further improve the recognition accuracy, and we can also setup specific keyword/key phrase alerts that highlight text passages with these words or phrases or quickly send you alert mails when these words or phrases are heard.

Consulting for Third Party Speech Products

We can help you boosting the experience with voice software by other vendors. Our team has successfully customized and deployed solutions for e.g. Nuance™’s Vocon™ suite, Amazon™’s Alexa Voice Service™ (used in the Echo™ devices) or the Pepper™ robots from Softbank™. Whether you need professional help with voice products as a voice newbie, or your product already has a voice control component with poor user feedback, let us offer you our professional support to make voice work in your product. Details here

Flexible Environments

All services are available on both embedded and connected environments; as SpeechFoundry is the only commercial voice software that uses the same voice core engin and same acoustic and language models in all environments, you have full flexibility to switch to a different environment later in your product cycle. E.g. imagine you start with a server-based voice solution because it is easy to setup, but consider then switching to an embedded version directly on your device for security reasons.

Embedded Environments

Your customized voice solution comes as an SDK that runs on your specific platform and in your environment. It interacts with your software via APIs and provides results as plain JSON messages.

Connected Environments

Your customized voice control runs on a remote server, so the devices that will use voice control need to be internet-enabled. Audio and result JSON messages are exchanged via specific internet protocols. We can set up dedicated voice servers in your server farm, or you can use Inferret’s voice servers.