bzikst/voicee-acestep15-xl-turbo

ACE-Step 1.5 XL Turbo for Voicee. Russian and multilingual song generation from lyrics and a style prompt.

Public
11 runs

Run bzikst/voicee-acestep15-xl-turbo with an API

Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.

Input schema

The fields you can use to run this model with an API. If you don't give a value for a field its default value will be used.

Field Type Default value Description
instructions
string
pop song, melodic vocals
Style caption for the song (genre, instruments, mood, BPM)
lyrics
string
Full lyrics with [Verse]/[Chorus] structure tags
description
string
Free-form song description for Simple mode (alternative to instructions+lyrics)
duration
integer
100

Max: 600

Audio duration in seconds (0 = automatic)
bpm
integer
0

Max: 300

Beats per minute (0 = auto)
seed
integer
-1
Random seed (-1 = random)
audio_format
None
mp3
Output audio format
vocal_language
string
ru
Vocal language
thinking
boolean
True
Let the 5Hz LM plan the song (slower, better structure)
use_cot_caption
boolean
False
Let the LM rewrite the style caption (may flatten the genre)
use_cot_language
boolean
False
Let the LM decide the vocal language
batch_size
integer
1

Min: 1

Max: 4

Number of variants to generate

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{
  "type": "string",
  "title": "Output",
  "format": "uri"
}