Search results

  1. HoodyP39

    【Kevin】 Family of Me: Extended Edit (from Over the Hedge)【SynthV Cover】 + SVP

    Read the description of the vid for SVP, credits and other tidbits about my life!
  2. HoodyP39

    Does anyone know any good engine for making your own text to speech in english?

    You don't really have to use Arpabet for training (which means you can just use normal sentences/alphabets, though Arpabet conversion is optional, but for that, it's best to use Tacotron2 forks that has Arpabet conversion if you wanted to use Arpabet for a slightly more accurate prononciation)...
  3. HoodyP39

    Does anyone know any good engine for making your own text to speech in english?

    Hey guys, as a Tacotron2 expert (and also part of the Uberduck contributors...). In order to create a really decent voice that covers most phonemes, I suggest doing at least about at least 30 minutes of speech data (total wavs duration), the best will be about 1-2 hours of speech data (the more...
  4. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Pretty odd... I'm not sure what is the issue here exactly but it's most likely because of either 2 things: 1. Your PC is installed as Windows 11 which might have some issues regarding programs (TALQu would most likely work well with Windows 8 - Windows 10) 2. Make sure you have installed .NET...
  5. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    UPDATE 2/14/2022: Haruqa announced a new version of TALQu has released and now the vocoder that was included in the current HaruqaCorpus became the default vocoder for TALQu! (Also model switching speeds have been increased and shortened!) Please update your TALQu for this new function!
  6. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Repost from VocaVerse for CUDA patch notes... When you were installing the CUDA 11.0 patch by running the bat file, does something like this appear in your CMD window? If so... DON'T CLOSE IMMEDIATELY!! It's just a warning regarding the fact that the Python library wasn't placed into PATH (which...
  7. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Notes (2/9/2022): So, if your TALQu was working fine until one day it just randomly decided to not load the GUI and come up with an error, first, close the opened TALQu if you still had it open (use Task Manager), back up a copy of your "Models" folder, delete ALL of the contents in the "TALQu"...
  8. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Well, I just sent an UTAForums conversation message about my Discord and Twitter contact... Feel free to DM me there because I will need more details about how you process the wavs and others before and during the Colab execution so that I have a clear understanding 1644217221 Notes to be taken...
  9. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Looks like it doesn't recognize the number of samples used... Well, if it can't recognize the files, I think you'll have to run the " 学習を再開する場合、もしくは、↑のノート上の前処理の実行を飛ばした場合、下のチェックを入れて、左の再生アイコンをクリックしてください" cell (aka the preprocessing wavs cell), to ensure that it processed the wavs into a format to...
  10. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Slight notes after making a TALQu voice... So, I just wanted to update some stuff in the process of making a TALQu model (once again as a reply bc of UTAForums edit limits...) 1. Just to clarify about the 5th cell in the training colab (前処理の確定と、アップロードした収録ファイルzipを前処理済みファイルzipで置き換えを行います), I'm a...
  11. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Well, I just made a TALQu voice! It's not complete yet (cuz I've only done like 174 lines in the corpus, my goal is to reach 1000 lines) but yeah! It's here lol
  12. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    UPDATE ON INSTALLING THE CUDA 11.0 PATCH: So, I just realized that the way to install it that I said previously was wrong actually. But here is the CORRECT way to install it... Extract the contents of the CUDA 11.0 patch into the installed TALQu folder and double click on the...
  13. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Hey guys, I can't really reedit bc of UTAForums limits... But I have been informed that Haruqa updated the Haruqa Corpus to now include a newer version of the vocoder (HaruqaOnly_軽量汎用_TALQu_HiFiGANModel.pt) in the model set up, and from what I've seen, it seemed to be better than the old one...
  14. HoodyP39

    Ehggs turts. I wannah et ehggs turts

    Ehggs turts. I wannah et ehggs turts
  15. HoodyP39

    How to create a TALQu voice?

    Hey guys! I have made an entire unofficial English guide for TALQu, which also includes voice model (TALQu's "voicebanks") creation! You can check it out here! https://utaforum.net/threads/talqu-an-unofficial-english-guide-thread-on-talqu-and-its-voice-model-creation.23552/
  16. HoodyP39

    List Of Vocal Synthesizers

    Let's add TALQu onto the list shall we?
  17. HoodyP39

    TALQu - An Unofficial English guide thread on TALQu and it's voice model creation.

    Hey guys, HoodyP here! Recently, on 24th of December 2021, Haruqa (creator of TALQu) initiated the "Lightning Project Challenge" where the goal of the challenge was for people to create a TALQu model within a week. The point of the challenge, according to Haruqa, was to in fact, "reach a...