Industry-leading speech-to-text technology with exceptional accuracy. Our company needed on premise solution, which was very easy to install using VM or containers. We had specific needs of recognition of names which were efficiently mitigated by usage of specific dictionaries. We only used the real-time speech to text capabilities for interractive and robotic applications. Review collected by and hosted on G2.com.
Real-time speech recognition applications still requires echo-cancellation microphones when using TTS at the same time to work efficiently and we would appreciate effort in "cocktail party" situations where multiple speakers speek at the same time. A first step in this direction is for sure Speechmatics diarization algorithm, that allows tracking and recognition of multiple speakers. Review collected by and hosted on G2.com.
