📚 Explore Now!
🌐 Dataset: huggingface.co/datasets/amphi…
📄 Paper: arxiv.org/abs/2501.15907
English
Amphion
59 posts
Amphion
@realamphion
Amphion is a toolkit for Audio, Music, and Speech Generation. GitHub: https://t.co/SDsdCO4M29 HF: https://t.co/3VJSoumiFS
Katılım Aralık 2023
54 Takip Edilen388 Takipçiler
✨ What’s New?
- 2x Scale: Expanded the original Emilia dataset from 101K to 200K+ hours with the new Emilia-YODAS dataset.
- Low-Resource Boost: Enhanced support for languages like German, French, and Japanese.
- Commercial Use: Emilia-YODAS is released under CC-BY
Amphion@realamphion
🚀Introducing Emilia-Large: 200K+ Hours of Open-Source Speech Data! We’re excited to release Emilia-Large, the largest TTS pretraining datasets! With 200K+ hours of multilingual speech data, fully open-source. It is ready to use for #TTS and #SpeechLM.
English
@234Sagyboy @discord @discordbots @DiscordBotDevs @_akhaliq @huggingface @reach_vb @Gradio Thanks for the feedback
English
@realamphion @discord @discordbots @DiscordBotDevs @_akhaliq @huggingface @reach_vb @Gradio I wish to provide some additional feedback I would appreciate if @realamphion team can direct me but if there is a feedback I were to give it would be 1) Speech to speech 2) Features like the one shown in image or like those microsoft Elate &EmoCtrl-TTS have Thanks
English
🚀🚀🚀 MaskGCT!
In addition to the HuggingFace demo: huggingface.co/spaces/amphion…
you can also join the discord space to play: discord.gg/fXmDZFur
Also pre-generated samples: maskgct.github.io
@discord @discordbots @DiscordBotDevs @_akhaliq
English
Amphion retweetledi
Sorry to interrupt but, YES, MaskGCT TTS works for French language !
I have not tested with other latin languages yet, but my guess is that it should work too 🤗
Vaibhav (VB) Srivastav@reach_vb
Fuck yeah! MaskGCT - New open SoTA Text to Speech model! 🔥 > Zero-shot voice cloning > Emotional TTS > Trained on 100K hours of data > Long form synthesis > Variable speed synthesis > Bilingual - Chinese & English > Available on Hugging Face Fully non-autoregressive architecture: > Stage 1: Predicts semantic tokens from text, using tokens extracted from a speech self-supervised learning (SSL) model > Stage 2: Predicts acoustic tokens conditioned on the semantic tokens. Synthesised: "Would you guys personally like to have a fake fireplace, an electric one, in your house? Or would you rather have a real fireplace? Let me know down below. Okay everybody, that's all for today's video and I hope you guys learned a bunch of furniture vocabulary!" TTS scene keeps getting lit! 🐐
English
Amphion retweetledi
I've added the MaskGCT TTS @gradio API to the Echo Mimic Space, so you can directly clone your voice before generating portrait generation 🤗
Try it —› huggingface.co/spaces/fffilon…
Vaibhav (VB) Srivastav@reach_vb
Fuck yeah! MaskGCT - New open SoTA Text to Speech model! 🔥 > Zero-shot voice cloning > Emotional TTS > Trained on 100K hours of data > Long form synthesis > Variable speed synthesis > Bilingual - Chinese & English > Available on Hugging Face Fully non-autoregressive architecture: > Stage 1: Predicts semantic tokens from text, using tokens extracted from a speech self-supervised learning (SSL) model > Stage 2: Predicts acoustic tokens conditioned on the semantic tokens. Synthesised: "Would you guys personally like to have a fake fireplace, an electric one, in your house? Or would you rather have a real fireplace? Let me know down below. Okay everybody, that's all for today's video and I hope you guys learned a bunch of furniture vocabulary!" TTS scene keeps getting lit! 🐐
English
GitHub: github.com/open-mmlab/Amp…
MaskGCT: github.com/open-mmlab/Amp…
Indonesia
🔥🔥🔥MaskGCT is hot, making Amphion on the list of GitHub Trending again!
> SoTA TTS model
> Zero-shot cloning
> Emotional TTS
> Multilingual, now supporting English and Chinese
> Fully non-autoregressive and duration controllable
Try in HF and discord.gg/fRaQpH7s
English
Fuck yeah! MaskGCT - New open SoTA Text to Speech model! 🔥
> Zero-shot voice cloning
> Emotional TTS
> Trained on 100K hours of data
> Long form synthesis
> Variable speed synthesis
> Bilingual - Chinese & English
> Available on Hugging Face
Fully non-autoregressive architecture:
> Stage 1: Predicts semantic tokens from text, using tokens extracted from a speech self-supervised learning (SSL) model
> Stage 2: Predicts acoustic tokens conditioned on the semantic tokens.
Synthesised: "Would you guys personally like to have a fake fireplace, an electric one, in your house? Or would you rather have a real fireplace? Let me know down below. Okay everybody, that's all for today's video and I hope you guys learned a bunch of furniture vocabulary!"
TTS scene keeps getting lit! 🐐
English
YouTube
English
@realamphion In speech conversion, where does the target speaker's timbre information come from？
English
🚀🚀🚀 A Zero-Shot TTS model MaskGCT (Masked Generative Codec Transformer) is open-sourced in Amphion now. Trained with Emilia. Only needs 5 sec speech to clone
Paper: arxiv.org/abs/2409.00750#
HF: huggingface.co/spaces/amphion…
Discord: discord.gg/fRaQpH7s
Watch the demo by MaskGCT
English
@realamphion Can you release the training /finetuning scripts & instructions, so that we can train on indian language (Telugu).
English
@realamphion Does it support podcast like conversations? I.e more than 2 speakers?
English
@mohamed17381489 Yes, it supports multi-lingual. We are going to release another checkpoint that supports 6 languages soon.
English
@realamphion YEEEEES!!!!! Thank you...Does it support multilingual? can we train it to speak Arabic?
English