Muvize v1.8 — Prompter, AI Transcription & Streaming Tiers

This release brings a brand new creative tool, smarter karaoke lyrics, and fairer streaming pricing. Here is what changed.

Prompter

There is a new tool in the sidebar called Prompter.

Describe your song idea in a sentence or two — the mood, setting, genre, whatever comes to mind — and Prompter generates a complete, platform-ready prompt you can paste directly into Suno, Udio, MusicGen, or Stable Audio.

Each platform has its own quirks. Suno wants short style tags and structured lyrics with section markers. Udio merges everything into a single prompt field. MusicGen needs a descriptive paragraph with BPM and instrumentation details. Prompter handles all of that for you, formatting the output for whichever platform you pick.

Four generation templates give you flexibility depending on how deep you want to go. Quick Spark keeps it simple — style tags plus singable lyrics from a single sentence. Full Pipeline adds production notes, title suggestions, and tags. Multi-Model focuses on technical parameters for instrumental generators. Suno v5 Pro is optimized for power users who want fine-tuned token usage and My Taste tags.

Prompter is multilingual. Write your idea in any language and the generated lyrics will match. The examples in the UI reflect this — you will see prompts in English, Indonesian, Japanese, and more.

Each generation costs 3 credits.

AI Lyrics Transcription

The karaoke maker got smarter about finding lyrics.

Previously, if Spotify did not have synced lyrics for a track, the karaoke feature had nothing to work with. Now there is a fallback. When Spotify lyrics are unavailable, Muvize automatically transcribes the vocal stem using AI. The result is converted to timed subtitles and added to your timeline just like Spotify lyrics would be.

Long lines are automatically split into readable chunks so nothing overflows on screen. The whole process is seamless — you do not need to do anything different. If lyrics exist on Spotify, those are used. If not, AI transcription kicks in.

This also works as a standalone feature in the studio transcription panel if you want to transcribe any audio clip directly.

Better Stem Separation

The engine behind stem splitting and karaoke has been upgraded.

The new model produces cleaner vocal isolation and instrumentals, especially on tracks with dense arrangements. It also runs with fewer parameters, which means faster processing on average. You do not need to change anything — every stem split and karaoke render now uses the improved engine automatically.