Video yükleniyor...
Video Yüklenemedi
NEW: Kokoro 82M - APACHE 2.0 licensed, Text to Speech model, trained on < 100 hours of audio 🔥
330,034 görüntüleme • 1 yıl önce •via X (Twitter)
11 Yorum

Vaibhav (VB) Srivastav1 yıl önce
Model checkpoint + voice pack here:

Vaibhav (VB) Srivastav1 yıl önce
We also added it to TTS Arena, go play with it now:

The Information1 yıl önce
The best new tech and business podcasts of 2024. Plus, podcasting's $400 million man.

mrfakename1 yıl önce
Ranking #4 on the TTS Arena above some commercial models 🤯

Vaibhav (VB) Srivastav1 yıl önce
🔥🔥🔥

Cavit Erginsoy1 yıl önce
did you say <100 hours?!

Vaibhav (VB) Srivastav1 yıl önce
yes! synthetic data ftw!

Dyson Dunbar1 yıl önce
its the first model that has convinced me to switch from xTTS because of the quality. It also helps that its about 10x faster as well on my 3090

Berat1 yıl önce
<100 hours of audio with 20 epoch. grokking for the win

Vaibhav (VB) Srivastav1 yıl önce
next step should be to fine-tune on Emilia

Pranjal Soni1 yıl önce
👀
