Langraf-SInDICATOR

Hardware and software for speech recognition and analysis. Processing of voice information and automatic recognition of free speech into text using artificial intelligence and machine engagement technologies.

Basic software modules:

DRV-SAAI-NOISE - Noise cleaning. Noise and background noise cleaning module: background music, noise, road noise. DRV-SAAI-DIAR - Speaker separation (diarization) and speaker recognition. Marking up speakers, restoring replicas and dialogs. DRV-SAAI-LANG - Identification of speech language. Classification of Common Speech Languages: major European languages (German, English, French, Spanish). Chinese, Arabic, Turkish are possible. DRV-SAAI-TRANS - Speech Transcripting. Speech to text translation. Wide range of input data: call center records (telephone conversations), video interviews, microphone records, and files from recorders. Multilingual communication management is also possible. DRV-SAAI-EVAL - Text analysis. Identification of keywords, main text subjects, identifying named entities. DRV-SAAI-LEARN - Training. Improving the quality of recognition by training the program on arrays of user records.