Latest mos study results for tts

[JRMeyer]

>>> erogol
[July 17, 2019, 9:53am]

Based on our latest MOS study, TTS is able to outperform any other
solution given a clear dataset. slash
Model details:

Judy Wave1: Tacotron + WaveRNN slash
Judy Wave2: Tacotron2 + WaveRNN slash
Judy GL1 : Tacotron + Griffin Lim slash
Judy GL2 : Tacotron2 + Griffin Lim

[This is an archived TTS discussion thread from discourse.mozilla.org/t/latest-mos-study-results-for-tts]

[JRMeyer]

>>> geneing
[August 5, 2019, 6:14pm]

What is the 'Judy' dataset? Is it publicly available?

Great result, by the way. Interesting that Taco1 outperforms Taco2 in
'Good' category but has more 'poor' results. Otherwise, they look on
par.

Is 'Judy' Taco1 pretrained model available?

[Archived Post]

[JRMeyer]

>>> erogol
[August 13, 2019, 8:19am]

Judy is an internal dataset 25 hours long. The only difference is that
it is recoded in better condition than LJSpeech.

[Archived Post]

[JRMeyer]

>>> geneing
[August 16, 2019, 5:19pm]

Is the dataset created by Mozilla? Any chance it could be shared with
the community?

[Archived Post]

[JRMeyer]

>>> erogol
[August 16, 2019, 6:15pm]

I don't think that would be possible in near future. Not because we
don't like but it is just too expensive with the current hype and its
recognition by the voice talents.

[Archived Post]