Skip to content

Indonesian

Acoustic Models

Model SR (kHz) Tokenizer Dataset
FastSpeech2 ID 22.05 Character-based Azure + WaveNet + Weildan
FastSpeech2 MFA ID v4 44.10 g2p_id (IPA) Weildan
FastSpeech2 MFA ID v5 44.10 g2p_id (IPA) Weildan (Mastered)
FastSpeech2 MFA ID v7 44.10 g2p_id (IPA) Azure
LightSpeech MFA ID 44.10 g2p_id (IPA) Azure
LightSpeech MFA ID v2 22.05 g2p_id (IPA) Azure
LightSpeech MFA ID v3 32.00 g2p_id (IPA) Azure
LightSpeech MFA ID v5 44.10 g2p_id (IPA) Althaf
Tacotron2 ID 22.05 Character-based Azure
Tacotron2 ID v2 44.10 Character-based Azure

Vocoder Models

Model SR (kHz) Dataset
MB-MelGAN HiFi PostNets ID 22.05 Azure + WaveNet + Weildan
MB-MelGAN HiFi PostNets ID v4 44.10 Weildan
MB-MelGAN HiFi PostNets ID v5 44.10 Weildan (Mastered)
MB-MelGAN HiFi PostNets ID v7 44.10 Azure
MB-MelGAN HiFi PostNets ID v8 44.10 Azure
MB-MelGAN HiFi PostNets ID v9 22.05 Azure
MB-MelGAN HiFi PostNets ID v10 32.00 Azure
MB-MelGAN HiFi PostNets ID v12 44.10 Althaf