Skip to content

Releases: CherokeeLanguage/IMS-Toucan

Models and code updated to allow speech rate adjustment.

19 Jul 03:13
@michael-conrad michael-conrad
2022.07.18.01
b4145df
Compare
Choose a tag to compare
Models and code updated to allow speech rate adjustment. Latest
Latest

Models attached.

Code updated to allow speech rate adjustment.

  • Clone the repo. Run git checkout 2022.07.18.01.
  • Using conda from miniconda setup the needed python environment: conda env create -f environment.yml
  • Copy Aligner-aligner.pt to Models/Aligner/aligner.pt.
  • Copy FastSpeech2_Cherokee_West-best.pt to Models/FastSpeech2_Cherokee_West/best.pt.
  • Copy HiFiGAN_combined-best.pt to Models/HiFiGAN_combined/best.pt.
  • Unzip IMS-Toucan-Reference-Voices.zip and place wav files into ref/.

Try using tts.py or look at the various run_chr*py scripts to see how to interface with library.

Full Changelog: 2022.05.12.00...2022.07.18.01

Assets 6
Loading