APITry the API →
HermesHermes Agent Docs
API
Back to News
modelAudioOpen Source

Antalia-2 Mini Beats 2.38B Models on Turkish Speech at Just 30MB

A 7.6M-parameter Turkish text-to-speech model runs 50x faster than real time on a laptop CPU, with the fewest character errors among 15 Turkish systems benchmarked.

Antalia-2 Mini Beats 2.38B Models on Turkish Speech at Just 30MB
Source
AlphaSignal
Published
Author
AlphaSignal Newsroom
Read
1 min read

A 7.6M-parameter Turkish text-to-speech model runs 50x faster than real time on a laptop CPU, with the fewest character errors among 15 Turkish systems benchmarked.

Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.

Read original report

Next reads

ACE-Step 1.5 Lets Developers Generate Music Locally Without PyTorch
AlphaSignal · model

ACE-Step 1.5 Lets Developers Generate Music Locally Without PyTorch

EMA Lightning Beats ElevenLabs on Turkish Speech in Just 34 MB
AlphaSignal · model

EMA Lightning Beats ElevenLabs on Turkish Speech in Just 34 MB

Open-Source VieNeu-TTS v3 Turbo Runs 48 kHz Vietnamese Speech on CPUs
AlphaSignal · model

Open-Source VieNeu-TTS v3 Turbo Runs 48 kHz Vietnamese Speech on CPUs