Build with the most accurate, realistic, and cost-effective APIs for speech-to-text, text-to-speech, and voice agents. Available in real-time and batch, cloud and self-hosted.