AssemblyAI Releases TTS-1 Speech Synthesis API, Priced at $15 per Million Characters

According to Techub News, AssemblyAI has released its first text-to-speech model AssemblyAI-TTS-1, which is now fully open. This API is mainly aimed at voice agents in real-time telephone call scenarios, priced at $15 per million characters, with no annual contracts or concurrent tier purchases required. The model was trained using millions of hours of audio data, including real conversation data, and employs continuous audio representation to enhance realism and reduce the generation time of the first audio segment. This release includes a library of over 4,000 voices and real-time voice cloning functionality, which only requires a short recording and speaker permission confirmation. Initially, it supports six languages: English, Spanish, French, German, Italian, and Portuguese. Developers can integrate through three interfaces: HTTP API, WebSocket API, or an OpenAI-compatible routing. (MarkTechPost)