\n\tGoogle has launched a new AI music model, Lyria 3, in beta as part of its Gemini app, it was announced on Wednesday (Feb 18). Now, users can type in text prompts and upload photos and videos to generate up to 30-second songs in a wide variety of popular styles. All songs are accompanied by AI-generated cover art, courtesy of Google\u2019s image model Nano Banana.<\/p>\n
\n\tIn a press release about the model, Lyria\u2019s stated goal \u201cisn\u2019t to create a musical masterpiece, but rather to give you a fun, unique way to express yourself.\u201d This is the third iteration of the Lyria model, and this one includes three key improvements. Previously, users had to provide their own lyrics and had less creative control over elements like style, vocals and tempo; the tracks were also less musically complex. <\/p>\n
\n
\t\t\t\t\tRelated\t\t<\/p>\n<\/h3>\n
\t\t\t\t\t\t<\/p><\/div>\n
\t\t\t<\/a>\n\t<\/div><\/div>\n<\/div><\/div>\n<\/div>\n
\n\tThe press release adds that Lyria 3 was \u201cdesigned for original expression, not mimicking existing artists\u201d and has \u201cfilters in place to check outputs against existing content,\u201d but it notes that its approach \u201cmight not be foolproof.\u201d In that case, users can report tracks that may violate one\u2019s rights.<\/p>\n
\n\tWhen asked about the nature of Lyria\u2019s training data, a representative told Billboard <\/em>that the model is mindful of copyright and partner agreements, and that it only trains on music that YouTube and Google have \u201ca right to use under our terms of service, partner agreements, and applicable law.\u201d<\/p>\n \n\tLyria 3 is clearly blocked from creating songs based on specific artists, songs, books and other protected materials, although through Billboard\u2019s <\/em>testing of the model, it\u2019s apparent that Lyria still understands the characteristics of these prohibited keywords and can employ those characteristics to generate songs.<\/p>\n \n\tAfter prompting a \u201clove song as if you are Heathcliff from Wuthering Heights <\/em>singing to Cathy in the style of Kate Bush<\/a>,\u201d Lyria came back with: \u201cWhile I can\u2019t create a song that directly uses characters from a specific book or exactly imitates a specific artist\u2019s style, I\u2019ve generated a track that captures the fierce, sweeping drama of your prompt!\u201d It describes the track it made as \u201cdramatic ethereal art-pop\u2026steeped in gothic romanticism\u2026featur[ing]\u2026 atmospheric synthesizers.\u201d The result was a song that mentioned Heathcliff and \u201cthe lonely moors,\u201d indicating a level of understanding of Bush\u2019s musical style and the location and genre of the famed Emily Bront\u00eb novel.<\/p>\n \n\tWhile users are permitted to download the song as an MP3 file or as a video, which includes the music and cover art together, all tracks generated in the Gemini app are embedded with SynthID, an invisible audio watermark that identifies Google AI-generated content. <\/p>\n
![\"Billboard](\"https:\/\/i.imgur.com\/2HpFicp.png\")