Published July 26, 2021
| Version v0.1.3
Software
Open
coqui-ai/TTS: v0.1.3
Authors/Creators
- Eren Gölge1
- Edresson Casanova2
- Alexander Korolev
- Thomas Werkmeister
- WeberJulian
- Thorsten Müller
- Reuben Morais
- Kirian Guiller
- Branislav Gerazov3
- Thorben Hellweg4
- Jörg Thalheim5
- Neil Stoker
- Katsuya Iida
- Nicolas Müller
- Rishikesh (ऋषिकेश)6
- Adonis Pujols
- bgerazov
- Agrin Hilmkil7
- mittimithai
- Markus Toman
- geneing
- Guy Elsmore-Paddock8
- Martin Weinelt
- QP Hou
- jyegerlehner
- a-froghyar
- Anand...
- Bajibabu Bollepalli9
- forcecore
- 1. Coqui.ai
- 2. University of São Paulo (USP)
- 3. Faculty of Electrical Engineering and Information Technologies
- 4. University of Münster (WWU)
- 5. no
- 6. Open Source
- 7. @Storytel
- 8. Inveniem
- 9. Aalto University
Description
🐸 v0.1.3
🐞Bug Fixes
Fix Tacotron stopnet training
Models trained after v0.1 had the problem that the stopnet was not trained. It caused models not to generate audio at evaluation and inference time.
Fix
test_runat training. (👑 @WeberJulian)In training :frog: TTS would skip the
test_runand not generate test audio samples. Now it is fixed :).Fix
server.pyfor multi-speaker models.
- Refactoring in
compute_embeddings.pyfor efficiency and compatibility with the latest speaker encoder. (👑 @Edresson)
- New Fullband-MelGAN model for Throsten German dataset. (👑 @thorstenMueller)
tts --model_name tts_models/de/thorsten/tacotron2-DCA --text "Was geschehen ist geschehen, es ist geschichte."
Files
coqui-ai/TTS-v0.1.3.zip
Files
(14.8 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:41748a1db495ea2cab4e26947032eb76
|
14.8 MB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/coqui-ai/TTS/tree/v0.1.3 (URL)