ArchiVoice (@VoiceArchival) profile picture

ArchiVoice

@VoiceArchival

We are ArchiVoice, a Taiwanese management team for various vocal synth projects, run and managed by @NekroTheCorpse 我們是來自台灣的凍聲響世，將聲音凍結在此刻，為您唱出最美妙的旋律。

 Taichung City, Taiwan
github.com/archivoice
Joined June 2022
  • Tweets79
  • Following26
  • Followers35
4Media
Photo by ArchiVoice (@VoiceArchival)Photo by ArchiVoice (@VoiceArchival)Photo by ArchiVoice (@VoiceArchival)Photo by ArchiVoice (@VoiceArchival)
yumeship yumeshipper yelling UST: vyreedy
6
37
734
the call is coming from inside the house
Rishie-P 🦋 (@Rishhie) profile pictureRishie-P 🦋@RishhieOct 28
HOLD ON A DAMN MINUTE
12
first time getting tagged, wheeeeeee
KuroTrash_Art/Khaos 【VZ】 🇨🇱 VOCALOID (@misfitkha0s) profile pictureKuroTrash_Art/Khaos 【VZ】 🇨🇱 VOCALOID@misfitkha0sOct 27
Replying to @misfitkha0s
This voicebank is able to sing in Korean and Chinese thanks to: Chinese Database Extra by @VoiceArchival (35 mins) Korean Database Extra by @tlrpclock (35 mins) Chinese + Korean No-Native Extra by @misfitkha0s (20 mins)
1
18
Hii! Here are some samples of my work on DiffSinger, including training and labeling. Akari Nem DiffSinger! (Voicebank Commission) This voicebank contains +2 hours of data/7 Vocalmodes: (Standard・Power・Light・Soft・Whisper・Falsetto・Rap) #DiffSinger #Commission #VTuber
2
12
73
1,442
ArchiVoice retweeted
KAISKE/H3xagonP (@ikekaiske) profile picture
KAISKE/H3xagonP@ikekaiskeOct 26
Today I'd like to preset apollo's newest art for his diffsinger update, this amazing work was made by @EnjouAkuma , plz follow him and support his work too
3
18
70
1,037
Here are some features in PRINTmov Splice that not many users use. 1/4 PRINTmov Splice started as a video editor, so overlaying elements is one of its core features (blending options will be added in the future).
2
13
49
1,053
ArchiVoice retweeted
StAkira (@StAkira) profile picture
StAkira@StAkiraOct 8
OpenUtau 0.1.573 Beta openutau.com
Image from StAkira (@StAkira)
10
189
19
1,037
25,733
ArchiVoice retweeted
Naysan♤ (@NaysanMLo7) profile picture
Naysan♤@NaysanMLo7Oct 5
Been brewing up a vocoder and architecture update. We are using auto-encoder and frequency domain neural vocoder instead. Music: Mili ustx: scoliosin once again mixing sucks but well this is a quick test
Naysan♤ (@NaysanMLo7) profile pictureNaysan♤@NaysanMLo7May 1
Using nsf hifigan on a tiny update with crosslang TEST Accent transferring implemented, but why would I want to hear Ritsu sing in English with strong Japanese accent lmao so im using English accent as a test music & svp: FLAVOR FOLEY mixing sucks but well this is a quick test
1
3
1
17
488
ArchiVoice retweeted
NeoDIVA.VS (@neodiva_vs) profile picture
NeoDIVA.VS@neodiva_vsSep 28
Hi Divas, WAVE 6 is HERE! Introducing 2YON, our newest REMIX vocalist voiced by StoneCapVA, as well as Zeraphys and Fuwa☆Fuwa! You can download all the new additions, plus updates to all previous NeoDIVA Vocalists, on our website! #NeoDIVA #DiffSinger #VocalSynth
Image from NeoDIVA.VS (@neodiva_vs)
5
25
6
57
2,284
これは公開情報やから、もう言うても大丈夫なんやけど、咲、後鬼、コタロー、狐面党の三人、スノーフレークの三人、で九人やね〜ξ😉ξ👍 人数多いから大変やけど、順次収録進めてるでー #ついなちゃん #DiffSinger
練星🍆 (@nerihoshi) profile picture練星🍆@nerihoshiSep 18
ついなちゃんメンバーのDiffSinger音源は、OpenUtauとかで使える音源になるのかな？ ついなちゃん以外との事だから、咲ちゃん達やスノフレ組、狐面党組あたりの中の何人かが使えるようになるっぽい？　＞RP
1
15
27
1,875
DiffSingerの音源は無料で使えるように公開するつもりやよ‼️ 全部で……、9キャラクターかな‼️ξ😉ξ👍 #ついなちゃん #DiffSinger
This Post is from a suspended account. Learn more
1
20
2
75
4,549
ArchiVoice retweeted
KAISKE/H3xagonP (@ikekaiske) profile picture
KAISKE/H3xagonP@ikekaiskeSep 18
Tsuina chan diffsinger???? 🫪 Omg
【鬼っ子ハンターついなちゃん】CV:門脇舞以様　節分マスコット🌟 @伝統文化&ヤングケアラー応援❣ (@Tuina_chan_PJ) profile picture【鬼っ子ハンターついなちゃん】CV:門脇舞以様　節分マスコット🌟 @伝統文化&ヤングケアラー応援❣
@Tuina_chan_PJSep 18
ウチ……ではないけど、実はDiffSinger音源は制作進行中やよ〜ξ😉ξ👍 しかも……複数が同時進行で収録中‼️‼️‼️ こちらも乞うご期待やで〜〜✨✨✨ξ😆ξ #ついなちゃん
4
7
79
2,512
If you’re an AI vocal synth development group working on ethically-made sound sources, I’d love to send about an hour of supplemental data in English, Tagalog, Spanish, and Japanese. Please DM if interested!
1
5
17
608
Chat, would you make a Chis-A x LIEE song please 🙏
Otomo Collective (@OtomoCollective) profile pictureOtomo Collective@OtomoCollectiveSep 12
Calling all creatives! We are looking for collaborators for our second Vocaloid album, Oshogatsu Otomo! Link to applications below.
2
20
598
ArchiVoice retweeted
Gianloop🍕 (@gnloop) profile picture
Gianloop🍕@gnloopSep 9
I've been informed my public dataset was also used for these ports, please be respectful about how you use community datasets!
ArchiVoice (@VoiceArchival) profile pictureArchiVoice@VoiceArchivalSep 9
It's come to our notice that some public datasets, including our own, ACV-001, have been used to train illegal diffsinger ports of commerical voices from other synth products, such as SynthesizerV, CeVIO, Voisona, and Vocaloid. Some of which have even been distributed.
Image from ArchiVoice (@VoiceArchival)
Image from ArchiVoice (@VoiceArchival)
8
1
31
1,344
近日發現似乎有人在利用一些包含我們的在內的公開資料庫訓練一些不大合法的二創diffsinger歌手模型，其中的違法資料來源包含SynthesizerV、CeVIO、Voisona、Vocaloid等。 其中一部分模型還被公開分享。
Image from ArchiVoice (@VoiceArchival)
Image from ArchiVoice (@VoiceArchival)
1
2
6
359
more replies
除此之外，mlem和hx這兩個音素也證實了ACV-001的使用，畢竟這兩者並不包含在我們發佈的音素表當中，排除其他使用我們音素表的其他資料庫的可能。 若有看到這種違法製作的二創模型，煩請幫忙檢舉或通知發佈者撤下來。
1
1
60
ACV資料庫都會繼續保持完全免費公開的狀態，但這些資料庫畢竟也是我們的心血，所以拜託別使用在這種地方，謝謝。
2
45
While we very much enjoy making and sharing our datasets, it should be noted that we do not, by any means, allow our datasets to be used for such illegal ports. The images above show the model includes Cantonese support, using a phoneme set that is currently only seen in ACV-001.
1
7
527
Aside from this, the "mlem" and "hx" phonemes are also a telltale sign of the ACV dataset's involvement in this. If you ever see an illegally made port, please fo report it and request it be taken down.
1
4
439
The ACV datasets will always remain public and free of use regardless, but please, don't use them for this purpose, for these datasets are essentially are blood and bone. Thank you
4
401
Load more