Lawrence John Haas

Voice message system using synthetic speech

A voice message service system stores subscriber-specific voice phrases (e.g., names) using digitally encoded phonemes while common system voice phrases are stored using digitally encoded voice samples. System voice announcements are formed by combining selected common system voice phrases with selected subscriber-specific voice phrases generated from phonemes. When a voice mail message is sent from one message system to another message system, the senders identification is sent using digitally encoded phonemes in the message header. The destination message system uses the received digitally encoded phonemes to generate the senders identification as a synthetic voice signal.