CereProc Integrations

CereProc is proud to announce CereWaveAI, our neural text-to speech system powered by advanced machine-learning technology. CereWave AI can be found in the CereVoice Cloud. CereWaveAI generates speech that sounds natural and more natural than any text-to-speech systems. It produces a new level inflection and emphasis that is human-like. The model creates audio waves from scratch using a deep neural network that was trained with large amounts of speech. The network learns how to create realistic speech waveforms by extracting the voice's structure during training. CereWave AI produces a voice almost identical to human speech. It also allows for full editing and control. You can change the voice to speak any language, gender or accent. While traditional text-to-speech systems can take 30 hours to record, CereWave AI requires only 4 hours to generate a high quality voice.