Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Pinnora's AI jingle generator turns a written sonic direction into brand music you can drop straight into an edit. Every run returns two takes of one hook, bright and grounded, with the lyric line held identical across both, so you are picking an arrangement rather than a second jingle. Both land as MP3s.
The tool is open above. This is the short version of what to do with it.
Name the genre, the lead timbre, the tempo feel and where the spot plays: warm acoustic guitar over brushed drums, 12s loopable hook, ends clean.
Duration runs 5 to 60 seconds and opens at 12. Eight to thirty is where a loop holds, and the number you set beats the model's own opinion.
One Mood from five, then up to two Instrumentation chips. BPM waits under Advanced at 60 to 180, and autofill never touches it.
Every run returns two takes of one hook, bright and grounded, with the lyric line identical across both. Download the MP3 or attach it to a video.
Short answers to what people ask before they start.
The Duration ruler runs from 5 to 60 seconds and opens at 12. The hint under it, 8 to 30 seconds is the loopable sweet spot, is the honest range: the music engine clamps any render to 30 seconds, so a 45 second pick still comes back as 30 seconds of audio. Short serves the goal anyway, since the mnemonic the brief is built around is two or three seconds long.
Two takes of the same song, labeled bright and grounded. The lyric hook is forced identical across both, word for word, so you are picking an arrangement rather than a second jingle. Each tab carries its own Music-gen prompt and Delivery notes naming the motif, key, dynamics and mix mood, plus an MP3 with a player and a download button.
Both. Before writing anything it pulls the project's brand profile, research, strategy, past winners and brand memory, and the brief step uses that to fix a sonic archetype and a two to three second motif before it settles tempo or instruments. Your Sonic direction leads and the brand material supports it. If that fetch fails or the project is empty, it falls back to the brand voice line and still renders.
Not usefully. The spec reaches the music engine as written, and the engine rejects anything reading as a copy of a real record. On that rejection the call retries once with the trigger phrasing cut: in the style of, sounds like, inspired by, reminiscent of, cover of, and anything inside quote marks. Genre, BPM, key, instrument names and lead timbre survive the cut, and they are what steers the sound.
The take still ships as text. You keep the Music-gen prompt, the Lyrics and the Delivery notes, the tab dot turns red instead of green, and the take shows the error it hit. A banner over the tabs counts how many of the two takes have audio. Attach to video only appears on a take that rendered, so re-run when you need that.
A run bills the brief that writes both specs, then one music clip per take, so two clips. Music is charged per clip rather than per second, which means a 30 second render costs the same as an 8 second one. Credits come out of your organization balance, and a take whose audio fails never reaches the charge, so it costs no clip.
The rest of the family. Every one is free to try and runs on the same engine.
