Google DeepMind launched Lyria 3.5 in Google Flow Music on Wednesday, delivering more expressive AI music generation with better vocals, sharper lyric coherence, and direct tempo control. The update is live now for all Google Flow Music users at no additional cost, marking the third significant Lyria milestone since February. According to Google's official announcement, the update touches four areas: richer melodic structures that sound more natural across a track, higher-quality lyrics with improved prompt adherence, more realistic and emotionally nuanced vocals with improved pronunciation, and more precise tempo and duration control. Lyria 3.5 outputs 48kHz, 16-bit PCM stereo audio. The underlying stack includes Lyria for audio generation, Gemini for natural language understanding, and Veo for visual components. ProducerAI, the startup Google acquired in February 2026, was built on the original Riffusion model, which Google replaced entirely with Lyria 3 at acquisition. The structural coherence improvements debuted in Lyria 3 Pro, which Lyria 3.5 now builds on, come from training the model to understand songs as str.