Meta's Voicebox Outperforms State-of-the-Art Models on Speech Synthesis

Meta recently announced Voicebox, a speech generation model that can perform text-to-speech (TTS) synthesis in six languages, as well as edit and remove noise from speech recordings. Voicebox is trained on over 50k hours of audio data and outperforms previous state-of-the-art models on several TTS benchmarks.

ai anthony art audio benchmarks data deep learning edit languages meta ml & data engineering natural language processing neural networks noise six speech speech generation state synthesis text text-to-speech tts voicebox