Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / lucidrains/spear-tts-pytorch issues and pull requests

#15 - EOS token not predicted while training from scratch

Issue - State: open - Opened by Kodhandarama 10 months ago - 3 comments

#14 - I have bad results of backtraslation

Issue - State: open - Opened by nassimabenammar about 1 year ago - 1 comment

#13 - Can I ask How to train tts model ?

Issue - State: open - Opened by dy2009 about 1 year ago

#11 - use cuda to generate audio - text data set

Pull Request - State: closed - Opened by KonstantineGoudz about 1 year ago - 1 comment

#10 - Inference for TTS

Issue - State: open - Opened by Wizard-The-Grey about 1 year ago - 1 comment

#9 - Change the `generate` code to check if a tensor, not specifically a torch.FloatTensor.

Pull Request - State: closed - Opened by itsjamie about 1 year ago - 13 comments

#8 - Fix an assert when running validation during training

Pull Request - State: closed - Opened by lucasnewman about 1 year ago - 1 comment

#7 - Minor trainer cleanup

Pull Request - State: closed - Opened by lucasnewman about 1 year ago

#6 - Add support for specifying a mask id during the speech-to-speech pretraining task

Pull Request - State: closed - Opened by lucasnewman about 1 year ago - 2 comments

#5 - Add a trainer for the text-to-semantic model

Pull Request - State: closed - Opened by lucasnewman over 1 year ago - 4 comments

#4 - Add trainers for the pretraining and backtranslation tasks

Pull Request - State: closed - Opened by lucasnewman over 1 year ago - 2 comments

#3 - Add a trainer & demo notebook for the speech-only pretraining task

Pull Request - State: closed - Opened by lucasnewman over 1 year ago - 9 comments

#2 - Questions about code

Issue - State: closed - Opened by shanhaidexiamo over 1 year ago - 9 comments