Mirelo SFX1.6 Video to Audio

Mirelo SFX 2.0

Generate synchronized soundtracks for a source video and return every video variation with sound. Preserves speech by default. Sound duration clamps to source length, up to 600 seconds.

P-Video Upscale

Upscale an existing video to 720p, 1080p, 2K, or 4K at 24, 30, or 60 FPS while preserving its aspect ratio and audio. Resolution targets use pixel area; larger sources are not downscaled. Turbo processes faster with a quality tradeoff at the same price. Accepts MP4 or MOV up to 512 MiB. Longer clips are trimmed to the beginning: at 24 FPS, about 100/50/25/12.4 seconds for 720p/1080p/2K/4K; limits shorten at higher FPS. Billing rounds the processed duration up to a whole second.

VEED Dubbing

Translate talking videos into 86 languages with cloned voices and synchronized lip movements. Detect the source language automatically or choose it, and supply names and key terms to preserve. Regional language tags use the base language without a regional accent. Source videos must contain speech and be at most 10 minutes long.

VEED Dubbing Pro

Translate talking videos into 94 languages with 14 regional accent options, cloned voices and synchronized lip movements. Supports multiple speakers and preserves background audio. Adjust voice cloning strength from 0 to 10 to balance the original voice and accent against naturalness in the target language. Detect the source language automatically or choose it, and supply names and key terms to preserve. Source videos must contain speech and be at most 10 minutes long.

MiniMax H3 Highspeed

MiniMax H3 Highspeed creates 4–15 second videos at 768p with native audio. Generate from text, animate a starting image with an optional end frame, or combine up to nine image, three video, and three audio references. Reference clips must be 2–15 seconds each, totaling at most 15 seconds per media type. Audio references require an image or video. Frame guidance and reference inputs cannot be combined.

MiniMax H3 Lite

MiniMax H3 Lite creates 4–15 second videos at 480p or 768p with native audio. Generate from text, animate a starting image with an optional end frame, or combine up to nine image, three video, and three audio references. Reference clips must be 2–15 seconds each, totaling at most 15 seconds per media type. Audio references require an image or video. Frame guidance and reference inputs cannot be combined.