AI Speech Recognition
For Conversational IVR, voice-enabled Chatbots, contact center calls transcription and other audio transcription requirements of an Enterprise.
Word Error Rate as low as the error rate of human transcriptionists. Whether deployed for IVR or a Chatbot, Voiceweb speech models will deliver as low as 5% Word Error Rate for the English language, and 6% WER for most of the other supported languages, even for narrowband audios with noise (IVR and Contact Center audios).
Easy to customize using contact center calls
to cover terminology or product names and maximize the accuracy (Word Error Rate)
For Cloud or on-premise deployment
Achieving highest self-service rate and customer satisfaction by correctly understanding the users’ request the first time, without error or retries. By maximizing the self-service rate, an Enterprise will improve multiple Contact Center metrics (AHT, hold time for callers, SLA etc.) and greatly decrease the Contact Center’s operating costs.
MRCP connector for direct integration with existing IVR platforms (e.g. Avaya, Genesys, Cisco and others) and direct replacement of pre-existing ASR engines such as the on-premise Nuance Recognizer
gRPC or Websockets for integration to cloud Contact Centers and other systems (e.g. chatbots)