HMM/DNN-based Speech Synthesis System (HTS) - Voice Demos

Voice Demos †

Festival Online Demo †

HTS voices can be used on Festival. To listen to speech samples, choose "???_arctic_hts" from the "Voice" pull-down list on Festvox on-line demo page. To listen to speech samples of the HTS-2007 and Nitech-HTS 2005 systems for the Blizzard Challenge 2007 and 2005, please visit CSTR on-line demo page.

Flite+hts_engine Demonstration Page †

HTS voices for English can be used on Flite+hts_engine. An on-line demo can be found at Flite+hts_engine Demonstration Page.

Open JTalk Demonstration Page †

HTS voices for Japanese can be used on Open JTalk. An on-line demo can be found at Open JTalk Demonstration Page.

Sinsy †

HMM-based singing voice synthesis system for Japanese, English, and Chinese (Mandarin) can be used on Sinsy.

Other HTS Voices †

• Nitech-NAIST English HTS voices for the Blizzard Challenge 2006 link
• Nitech-HTS English voices for the Blizzard Challenge 2005 link
• Nitech-HTS Japanese voices link
• Japanese HTS singing voice at Interspeech 2006 by Keijiro Saino, Nitech pdf
• Brazilian and European Portuguese HTS voices by Ranniery Maia, Nitech, Universidade do Porto, and UFRJ link
• Angry Japanese HTS voice by Ryosuke Tsuzuki, Nitech ppt
• Korean HTS voice by Sang-Jin Kim, Information & Commun. Univ. link
• HTS arbitrary speaker's voice, emotional expressions & speaking styles by TIT/CSTR link
• Xhosa, Zulu, and South Africal English HTS voices by Albert Visagie, Stellenbosch University link
• HTS voice on Inuits Dance link
• Catalan HTS voices by Antonio Bonafonte, Universitat Politcnica de Catalunya link
• HMM-based speech synthesis using a DSM of the residual signal link

Misc †

Talk about HMM-based speech synthesis

• Slides pdf (51 MB)
• Demos link