Speech recognition
The Automatic Speech Recognition (ASR) technology can:
— filter background noises,
— detect an end of utterance,
— detect speaker’s emotions,
— insert punctuation marks.
Speech may be recognized better and quicker with the use of hints. They help to understand the user’s speech at a specific moment and to accelerate the system response

Speech synthesis
The Text to Speech (TTS) technology generates the speech that sounds natural.
SmartSpeech provides various speakers’ voices. You can choose the timbre, tone and mood.
The TTS technology uses unique models for marking accents and pronouncing the letter ‘ё’, which allows decreasing the number of phonetic mistakes and inaccuracies.
It also uses a large text normalization package allowing to pronounce digits, addresses, names and other complex texts

Easy API integration
SmartSpeech API allows you to connect speech recognition and synthesis to your products and services quickly and efficiently.
API uses HTTP and gRPC protocols, and therefore it can be integrated with almost any system