Cloud, hybrid, embedded - we support you with state-of-the-art technologies in the area of speech recognition.
You decide, if you attach value to data protection and process speech data only locally. Or if you want to use services like Amazon Alexa, Microsoft Cortana or Goolge to implement the possibility for extensive queries.
With over 15 years experience, we can support you with all these technologies. Just contact us.
| Server-based / Cloud |
Hybrid (Cloud + Embedded) | Embedded | |
|---|---|---|---|
| Fits for | Unlimited vocabulary, broad subjects | Large vocabulary | Defined vocabulary |
| Recognizer | Local server or cloud | Combination of cloud and embedded | Embedded (locally integrated in a device) |
| Applications | Speech-to-Text, Text-to-Speech, LVCSR | NLU | Command and control |
| Languages | 79 languages and dialects | 30 languages and dialects | 30 languages and dialects |
| Our services | GUI-based modelling and design tool for speech dialogue development | Command definition, operating functions | Command definition, operating functions |
| Training of acoustic and language models | Grammar creation and/or SLM modelling | Grammar creation and/or SLM modelling | |
| Speaker and topic adaptation | Speech dialogue procedure, prompts, operating tones | Speech dialogue procedure, prompts, operating tones | |
| Enlargement of dictionaries, OOV recognition | Optimisation of robustness against operating errors / OOV | Optimisation of robustness against operating errors / OOV | |
| Modelling of word classes and categories | Transfer to more languages | Transfer to more languages | |
| Semantical processing and analysis of meaning in texts |