精选理由
Pika Labs 新出的 Pika Music 模型，能同时用文字、歌词、声音和音乐来生成一首歌，比分开处理更方便。
Pika Labs 发布了新的 Pika Music 模型。该模型接受文本、歌词、语音参考和音乐参考四种输入模态。这些信号在一个共享的潜在扩散解码器中融合，生成完整的音乐曲目。它不是一系列独立的工作流程。
原文 · Pika Labs
A closer look at our new Pika Music model, which accepts four input modalities: text, lyrics, voice ...
A closer look at our new Pika Music model, which accepts four input modalities: text, lyrics, voice references, and music references—individually or in combination. Those signals meet inside one shared latent-diffusion decoder, which resolves musical structure, vocal character, written direction, and reference “vibe” as it generates a complete track. Not a chain of separate workflows. Your browser does not support the video tag. 🔗 View on Twitter 💬 2 🔄 3 ❤️ 16 👀 1802 📊 5 ⚡