April 2026 Summaries

Mist v3 is the latest iteration of the text-to-speech (TTS) model designed to meet the latency and throughput needs of enterprise voice deployments by offering significantly faster processing and high reliability. While the voice characteristics remain the same as previous versions, the underlying infrastructure has been revamped to achieve approximately 40ms time-to-first-byte on specific hardware, enhancing the performance of real-time conversational applications. This upgrade addresses the bottleneck issues often faced in voice AI pipelines, particularly for high-volume operations like enterprise contact centers, by ensuring high throughput and maintaining pronunciation control along with SSML features. Mist v3's improvements have notably benefitted entities like Attune, a healthcare platform, which uses the model to maintain HIPAA compliance and deliver empathetic, rapid responses during patient interactions. The model is available for both cloud and on-premises deployments, offering flexibility and control over data and infrastructure to meet various enterprise needs.