On March 20, OpenAI, the pioneering U.S.-based research hub for open artificial intelligence, unveiled that its API would integrate advanced transcription and speech generation AI models, marking a significant improvement over previous iterations. Notably, the new text-to-speech model, gpt-4o-mini-tts, boasts more refined and lifelike speech effects, alongside enhanced controllability when compared to previous speech synthesis models. Developers now have the capability to effortlessly tailor the model's vocal style through intuitive natural language commands, catering to a diverse array of user requirements. This groundbreaking enhancement promises to deliver a more personalized and superior speech experience to end-users.
