Technologies

Voxpass sells web and mobile applications using voice and speaker recognition, in order to digitalize natural speech.  Without prior learning, without forcing participants to adapt their speech.

Voxpass Apis specifically support multi-speaker, multi-langage and includes features like:

  • voice analysis,
  • text mining,
  • translation.

With Voxpass, no need for end-users to spend time with learning phases : our solutions will adapt automaticaly to speakers and context.

One innovation is to mix several speech recognition engines and speaker identification techniques, in order to produce the best  transcription confidence rate.

Today there is no similar solution on the market.

Capture

Audio device and signal processing for better results using speech recognition

DSP/ML

Signal analysis and pattern identification using machine learning models

Transcription

Converting speech to text automatically

Diarisation

Speaker segmentation using voice features detection with machine learning models

Identification

Speaker identification using voice characteristics with machine learning models

Emotion

Detection of emotion based on the analysis of variations in the voice signal

Translation

Multilingual translation based on a machine learning algorithm that takes into account the semantic context

Summary

Text mining with strong semantic analysis

Analysis

Understanding of what has been said and triggering actions. Extraction of strong words of meaning

Feelings

Sense detection based on the semantic configuration of language elements

Topics

Extraction of topics covered in a conversation