Skip to content
Trending storyDeveloping

Ant Ling's Ming-3.0-tts-0.5B clones a voice locally on an RTX 3090

1 article1 sourcesince Nov 1Last article

Overview

AISummary of 1 article

Ant Ling says its Ming-3.0-tts-0.5B text-to-speech model runs locally on an RTX 3090 and generates about 4 seconds of speech in 2–3 seconds.

Tester AJ (@ItsmeAjayKV) reports it cloned his voice from a 5-second sample recorded in fan noise, and says the model uses about 3.3 GB of VRAM with a total disk size under 3 GB.

The speed and memory figures come from Ant Ling's announcement and AJ's post, and AJ's account is a single informal test. No independent benchmark of the model's output quality is reported.

Written by AI from the articles below · updated Nov 1, 10:31 PM ET

Check the sources:

Latest developmentNov 1, 10:24 PM ET
Ming-3.0-tts-0.5B clones a voice from a 5-second sample on an RTX 3090

Article timeline

The articles in this story. Times are ET.

Nov 1
  1. Ant LingOfficial
    Ming-3.0-tts-0.5B clones a voice from a 5-second sample on an RTX 3090

    AIAnt Ling reports that Ming-3.0-tts-0.5B runs locally on an RTX 3090, and tester @ItsmeAjayKV says it cloned his voice from a 5-second sample recorded in fan noise. He says the model uses about 3.3 GB of VRAM, is under 3 GB on disk, and generates about 4 seconds of speech in 2–3 seconds.

Heat Trend

Current comparable heat
6
Peak comparable heat
10Nov 1, 11:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.