黒木武史 (@KRTK_12) header
黒木武史 (@KRTK_12) profile picture

🌈人生难得一知己,千古知音最难求。

Null
Joined November 2015
Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)Photo by 黒木武史 (@KRTK_12)
HachiTune 2 For more updates, please follow our GitHub and official website. The first release of HachiTune 2 is expected to be available soon. hachitune.app #OpenSource #DAW #VocalProduction #PitchCorrection #AudioPlugin #VST
HachiTune 2 A free, open-source neural network–powered pitch correction tool is coming soon... #OpenSource #DAW #VocalProduction
15
2
103
12,760
YumeSampler is a resampler based on the NHV vocoder, capable of delivering output quality of up to 48 kHz / 24-bit. It may be open-sourced, along with its model weights, in the future. Demo UST by echoblossom123 and revolocities
1
81
7
448
10,745
ゴジラ、襲来。 ニャジラ、就寝。
🤖 Made with AI
Readers added context they thought people might want to know
AIで生成された画像です。 OpenAIが提供するチェッカーで、ツールを使用した事を意味するシグナル(SynthID)が検出されています。 openai.com/ja-JP/research…
1,973
86,847
2,296
678,420
16,477,049
继续打卡,顺便改开始减一点脂了💪
27
766
🥺打卡加油💪
4
396
💪🥺
1
1
23
1,070
在努力变大了增肌好痛苦🥺
1
5
558
黒木武史 retweeted
if we had Claude in 1998 😁
82
463
44
3,797
301,762
黒木武史 retweeted
This is a tool that uses the iPhone’s TrueDepth camera to capture facial motion data alongside synchronized audio. After weighing cost against data accuracy, we chose this approach as a low-cost yet precise solution for collecting training data. The project originated in early 2025 as an experimental offshoot of our work within the DiffSinger community. Once recording begins, the app simultaneously captures audio and temporally aligned facial motion data, with a particular focus on lip movements. These synchronized recordings are then used for model training. Through this approach, we eventually developed an inference model capable of generating corresponding lip-sync curves from audio alone. Although the application’s original mission has largely been accomplished, I recently revisited the project, made some updates, and submitted a new version to the App Store. The app runs entirely offline and never connects to the internet. All recorded data stays locally on your device. The project is also fully open source, so anyone can build the app themselves using the source code available on GitHub. Project repository: github.com/KCKT0112/LipsSync
1
414
黒木武史 retweeted
got bored and reverse engineered the HID APIs for Apple's unreleased touchscreen MacBooks made an iPhone duo remote client so I can mess with native touch support on macOS lol
35
36
8
699
41,293
Suck
#VOCALOID7 is coming November 11.
4
585
This is a tool that uses the iPhone’s TrueDepth camera to capture facial motion data alongside synchronized audio. After weighing cost against data accuracy, we chose this approach as a low-cost yet precise solution for collecting training data. The project originated in early 2025 as an experimental offshoot of our work within the DiffSinger community. Once recording begins, the app simultaneously captures audio and temporally aligned facial motion data, with a particular focus on lip movements. These synchronized recordings are then used for model training. Through this approach, we eventually developed an inference model capable of generating corresponding lip-sync curves from audio alone. Although the application’s original mission has largely been accomplished, I recently revisited the project, made some updates, and submitted a new version to the App Store. The app runs entirely offline and never connects to the internet. All recorded data stays locally on your device. The project is also fully open source, so anyone can build the app themselves using the source code available on GitHub. Project repository: github.com/KCKT0112/LipsSync
1
414
黒木武史 retweeted
新的歌声合成引擎
1
19
1
110
3,289
这个世界完蛋了(确信
完成しました。 【MV】イキスギ☆ハルシネーション
1
544
黒木武史 retweeted
😭😭 bruh
74
796
57
12,981
553,211
黒木武史 retweeted
Winner announcement in 24 hours! We'll randomly draw 100 eligible winners with help from @grok. Make sure you follow us, like & repost to be in the running. Good luck! 🍀
My iPhone Duo video went viral, so I'm giving away 100 iPhones on X 😛
13,408
16,874
944
24,607
1,556,645
黒木武史 retweeted
My iPhone Duo video went viral, so I'm giving away 100 iPhones on X 😛
91,600
92,230
6,535
139,617
6,180,132
黒木武史 retweeted
Some Fashion POP Everybody wants to be Julia💅 #ACEStudio
3
8
410
黒木武史 retweeted
开源小螃蟹机器人 Jumper 太酷了:22 个自由度,又萌又灵活,能搬东西,还能跳跃。 创作者已把项目开源,自己就能动手搭一个。 github.com/KingKongRobotics/…
316
1,199
177
8,319
522,545