YouTube is rolling out new podcast-oriented features for Premium subscribers, starting today on Android and coming later to iOS. The key addition is an on-the-go mode that shifts playback toward an audio-first layout, with larger simplified controls, a still image replacing video, and a timeline. It is a modest step toward making YouTube more comfortable for podcast listening, not a full podcast-app overhaul.
The piece highlights a trend in the Suno subreddit: users are not merely generating AI songs, but listening almost exclusively to their own outputs. Some reportedly say they have stopped using traditional streaming platforms and now spend their listening time on AI-made music. The article frames this less as a product update and more as cultural commentary on personalization, taste, and the social meaning of music.
Spotify 與環球音樂(UMG)合作推出全新 AI 混音工具,聲稱能讓「超級粉絲」自由創作喜愛歌手的翻唱與混音版本。然而,網路上早已充斥著大量低質量的 AI 翻唱歌曲(如雷鬼版《Smells Like Teen Spirit》)。評論指出,這項工具非但無法提升粉絲體驗,反而可能讓串流平台充斥更多單調、廉價的 AI 生成音樂,其商業與藝術價值令人質疑。
開發者 Mike Veerman 製作了一個名為「tokenspeed」的 HTML 模擬工具,能呈現大語言模型(LLM)在每秒 5 到 800 個 Token 之間的生成速度。當各大廠商宣稱其模型達到特定 Token 速度時,使用者常難以想像其實際體感。此工具能幫助開發者與設計師直觀評估不同速度下的使用者體驗與 UI 設計。