Overview

TwinCAT Speech enables people and machines to interact via speech input and output.

The interaction with TwinCAT takes place in different languages and can be used across industries in many applications from mechanical engineering to building automation. For example, the operating or maintenance staff can be informed about or influence the state of the controller.

TwinCAT Speech provides the following functions:

• Communication via the audio output of the PC-based controller (sound card with speakers, headset, etc.)
• Speech input and output in the languages supported by Microsoft SAPI.
• Speech output in the languages that Amazon Polly provides.
• Caching of generated speech outputs

Through the use of Microsoft SAPI and the caching of speech outputs that may be generated online, you can achieve offline use.

TwinCAT Speech is based on the two components Automatic Speech Recognition (ASR) and Text-to-Speech (TTS). ASR is the speech recognition and is responsible for ensuring that speech input, which is recorded via a microphone, is usable in the PLC. TTS is the text-to-speech output and enables the machine to output speech.

TwinCAT Speech has an online function that is provided with the aid of the Amazon text-to-speech service Polly. It is possible to set different voices and to cache the audio files generated online.

The speech input has an offline function, which is realized by the built-in functions of the Windows operating system.