PlayDialog Beta

Play.ai just launched PlayDialog Beta, a smart AI speech model that creates conversational podcast audio. This tool uses the history of conversations to adjust tone, emotion, and speech rate, making the voice synthesis sound more natural. It''s great for creating authentic conversational experiences like narration, voice dubbing, synthetic podcasts, and immersive one-on-one voice interactions in commercial settings.

Key Features

PlayDialog has several standout features.

Natural Sounding Speech: It controls prosody, intonation, emotion, and pacing to deliver speech that closely matches human conversation.

Advanced Architecture: The model includes an ''Adaptive Speech Contextualizer'' (ASC) that uses the full context and history of a conversation to enrich each response with appropriate prosody, tone, and emotion.

Training and Performance: Trained on hundreds of millions of real conversations, PlayDialog is about ten times larger than Play AI 3.0 mini and outperformed leading competitive models in blind tests, scoring highest in expressiveness.

Real-Time Applications: PlayDialog supports streaming from LLMs using websockets, enabling fast responses to LLM-powered applications.

Benefits

PlayDialog offers more natural and expressive speech, making it ideal for creating engaging and authentic conversational experiences.

Use Cases

PlayDialog can be used for a variety of applications.

Narration and Podcast Creation: Create powerful, natural-sounding narrations, podcasts, briefings, and more from multiple media types.

API Access: Developers can use PlayDialog and PlayNote through APIs to generate engaging content at a large scale.

Commercial Applications: Create immersive and engaging 1:1 voice experiences with customers in business contexts.

Funding

Play AI has raised $21 million in funding. This investment will be used to enhance its generative AI voice models and voice agent platform, aiming to deliver powerful, emotive, and human-like voice interfaces for any application.