AI finally sleeps, Gemini 4 Carbon, ETH cracked, energy-based models, Anthropic IPO: AI NEWS
0.6B open TTS with zero-shot voice cloning from a few seconds of audio, 23 languages, 2.6GB, Apache 2.
Compact 0.6B open text-to-speech model with zero-shot voice cloning and support for 23 languages.
Updated EU status checked Last shown
Audio8 TTS Realistic 0.6B at a glance
- What is Audio8 TTS Realistic 0.6B?
- Audio8 TTS Realistic 0.6B is a model. Compact 0.6B open text-to-speech model with zero-shot voice cloning and support for 23 languages.
- Does Audio8 TTS Realistic 0.6B work in the EU?
- Available in the EU. Reported as an Apache 2.0 open-weights model that can be downloaded and run anywhere, with no regional restriction found; the license claim is not confirmed on the fetched page. Last verified 14 Nov 2026.
- How much does Audio8 TTS Realistic 0.6B cost?
- Open source. Reported as open weights under Apache 2.0 (per a YouTube creator); not confirmed on the fetched page. Licence: Apache-2.0.
- Where was Audio8 TTS Realistic 0.6B shown?
- In 1 creator video by AI Search. Most recently by AI Search on 1 Nov 2026, at 38:26 in “AI finally sleeps, Gemini 4 Carbon, ETH cracked, energy-based models, Anthropic IPO: AI NEWS”.
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
About
Audio8-TTS-Realistic (listed on its demo page as Audio8-TTS-Realistic-Air) is a 0.6B-parameter text-to-speech model built on Audio8-TTS-Preview. According to the project page, it adds richer, more expressive speech and expands language support to 23 languages while keeping the model size at 0.6B.
The demo page compares samples of the same text and reference voice across Preview, Breeze-TTS-2, Qwen3-TTS and Higgs TTS 3, with 24 examples. The page states these are selected examples and not an overall ranking. A YouTube creator summary describes zero-shot voice cloning from a few seconds of reference audio, a roughly 2.6 GB download and an Apache 2.0 license; these details were not confirmed on the fetched page.
The model is intended for developers who want a small, locally runnable TTS model. The vendor is not identified in the provided material, and no repository, Hugging Face card or documentation links were found, so license and weight availability should be verified at the source.
Key features
- 0.6B parameters, about 2.6 GB reported size
- Zero-shot voice cloning from a few seconds of reference audio (per creator report)
- 23 supported languages
- More expressive speech than Audio8-TTS-Preview
- Reported Apache 2.0 license
1 moment
As seen on YouTube
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Keep exploring
Similar AI tools
Closest neighbours by meaning, not just by category.
AT
Audio8 TTS 0.1B
Model · Edge0
Open-weights 0.17B-parameter multilingual text-to-speech model with zero-shot voice cloning, runs on CPU via ONNX.
Available in EUVoice & Audio
19 Aug 2026Open source
DO
dots.tts
Model · Rednote
Open-source 2B-parameter zero-shot voice-cloning text-to-speech model from Rednote's hilab team.
Available in EUVoice & Audio
Release date n/aOpen source
IT
Index TTS 2.5
Model · IndexTeam
Open-weights zero-shot text-to-speech model from Bilibili's IndexTeam with voice cloning and emotion control in 5 languages.
Available in EUVoice & Audio
10 Aug 2026Open source
WA
WavTTS
Model · ByteDance
Research zero-shot TTS model that generates speech by modeling raw waveforms directly with flow-matching DiT.
UnverifiedVoice & Audio
Release date n/a
SA
Scenema Audio
Model
Expressive text-to-speech model with zero-shot voice cloning, derived from the LTX 2.3 audio model.
Available in EUVoice & Audio
Release date n/aFreemium
P1
Paradee 1.2
Model
8.5 MB, 8M-parameter text-to-speech model distilled from Kokoro-82M, with four preset voices and CPU inference.
Available in EUVoice & Audio
Release date n/aOpen source