Yurii Paniv
Yurii Paniv
### Describe the bug If you select dark mode on HuggingFace, text in `description` and `article` fields is not visible, because it's dark grey text on black background. ### Is...
```python3 vitsArgs = VitsArgs( # hifi V3 resblock_type_decoder = '2', upsample_rates_decoder = [8,8,4], upsample_kernel_sizes_decoder = [16,16,8], upsample_initial_channel_decoder = 256, resblock_kernel_sizes_decoder = [3,5,7], resblock_dilation_sizes_decoder = [[1,2], [2,6], [3,12]], ) ```
Priority list: - [ ] Select audio tempo https://github.com/robinhad/ukrainian-tts/issues/39 - [ ] Memory leak https://github.com/robinhad/ukrainian-tts/issues/30 - [ ] Handling of numbers https://github.com/robinhad/ukrainian-tts/issues/21 - [ ] New stress model https://github.com/robinhad/ukrainian-tts/issues/15 -...
Specify audio speed whether using audio speedup like [Python playback without chipmunk effect](https://medium.com/prog-ramming-solutions/python-music-playback-speed-1x-2x-3x-change-without-chipmunk-effect-890eb10826c1) or with model (like VITS, FastSpeech 2)
- [ ] Test available phonemizers: espeak, lang_uk - [ ] Should be portable - x86_64, Windows, ARM, Linux - [ ] Without assumptions or simplifications - output how a...
Running model on small sentences (130 characters), memory grows rapidly 4GB+. Target: constant memory usage, no more than 2 GB at least.
- [ ] Мало бути 2 людей -> Мало бути двоє людей - [ ] Car plate numbers - [ ] Вона народилась 1923 року -> Вона народилась тисяча дев'ятсот...
- [ ] stress model for words not present in dictionary - [ ] automatic stress based on sentence (text?) level context - [ ] benchmark and dataset with edge...
Кидайте у форматі: Мало бути 2 людей -> Мало бути двоє людей