Describe the music you want — genre, mood, instruments, tempo.
Remix Existing Track
Upload or paste a URL to an existing audio track, then describe how to remix it.
Paste URL
Upload File
Choose audio
Uploading…
file.mp3✕
Extend a Track
Continue an existing track with new sections. Upload an MP3 or paste a URL.
Paste URL
Upload MP3
Choose audio
Uploading…
track.mp3✕
Lyria Music Generation
Lyria 3 Clip — Short clips, loops, previews · 30 seconds · MP3 Lyria 3 Pro — Full-length songs with verses, choruses, bridges · MP3 / WAV
Describe genre, mood, tempo, instruments, and structural cues (e.g. "starts soft, builds to a climax").
Results
Generate a track to see results.
Lip-Sync Video to Audio
Sync-Lipsync — High-fidelity facial alignment LatentSync — Fast with smooth temporal consistency Creatify — Optimized for marketing/social formats Veed — Broad video format support
Paste URL
Upload File
Choose video
Uploading…
video.mp4✕
Paste URL
Upload File
Choose audio
Uploading…
audio.mp3✕
Results
Submit a lip-sync job to see results.
Text → Audio
Video → Audio
Text to Audio / Sound Effects
Generate Foley, ambient sound effects, or speech from a text description.
Video → Synchronized Audio
Generate audio that matches the motion and content of an existing video.
Upload File
Paste URL
Choose video
Uploading…
video.mp4✕ Remove
Results
Generate audio to see results.
Clone a Voice
Upload a short reference audio sample (10–30s works best). The model captures tone, accent, rhythm, and emotion — then speaks any text you provide in that voice.
Upload File
Paste URL
Choose audio
Uploading…
reference.wav✕ Remove
Min 8 characters, must start with a letter and include both letters and numbers.
Max 2000 characters. This text will be spoken in the cloned voice as a preview.
Cloned Voice Results
Each result is an audio preview of the cloned voice speaking your text. The voice ID can be reused for future inference with any Minimax model.
Clone a voice to hear the result.
01
Choose Voice
Select a character voice — 30 Google Gemini TTS voices