Lyria 3.5 Usage Map: Podcast Soundtrack Scenarios

Taking "a usage map for voiceover/soundtrack/podcast" as the topic, use Lyria 3.5 to generate an instrumental soundtrack track for a podcast episode: structurally, it enters with a low-volume bed first, gradually adds rhythmic and harmonic layers in the middle section, and returns to a simple keyboard motif at the end, making it easy to lower the volume when the host starts speaking and raise it during narrative segments. In terms of composition, this entry only uses the music description (prompt), title, length, bpm, seed, and other non-media input parameters declared in the author's pipeline, and does not involve any reference images or audio materials, so there is no material generation step. The prompt describes the genre, instruments, mood, and section structure following the approach of the official prompt guide, with the duration kept within a few minutes to match the model's "several minutes" scale of output characteristics. Applicable scenarios: opening/background music for podcasts and interview programs, voiceover background music for short videos, neutral-emotion beds for brand content, and prototype validation of custom soundtracks that need clear structure and easy cutting. Note that Lyria 3.5 is only available through the Gemini API, outputs watermarked MP3s, and results are not cacheable; watermark and licensing requirements should be confirmed before formal commercial use.