Hello π, nice that you're here π.
I'm a voice/TTS enthusiast sharing his voice with the community.
β 23 hours "normal" dataset
β "Emotional" dataset (ππ²π₯΄π€π₯±)
β Pretrained TTS models
β€ for a π on Github.
https://t.co/SdQRHQV458
π₯ Check out my new YouTube video! First impressions of open source TTS: Kokoro, Fish, Llasa & Spark. π€
Discover the latest in voice tech: https://t.co/02caTgx812
#opensource#tts#voicetech
"Dr. Smith paid $1,234.56 at 3:30pm"
TTS π€: has mental breakdownπ
Tired of your AI voice butchering numbers & abbreviations? Got a fix for that!
New tutorial: How to make ANY TTS (even good old espeak!) speak like a human β https://t.co/0ikSCGySgv
#TTS#TextToSpeech#VoiceAI
Open voice tech enthusiasts, rejoice! π Finding 'living room compatible' hw has always been a challengeβbut with @home_assistant Voice (Preview Edition), we're closer than ever! π
Check out my Thorsten-Voice YouTube channel for tutorials and tips π.
https://t.co/X5mxNqU2Nl
βΌοΈ Do you have a teen who codes in your life? Have them join Hack Club's High Seas!
We're excited to sponsor this challenge by providing Home Assistant Greens to buy with earned Doubloons! π
Find the details here ππΌ ends Jan 31!
https://t.co/AWJYdKYlyO
https://t.co/AWJYdKYlyO
Pst. You, yeah you.
Have you hit that π on our YouTube channel to get notifications? We're pretty sure there's something happening on Thursday... π§
Link in reply ππΌ
ποΈ Create your personal AI voice clone with just 10s of audio input!
I uploaded a step-by-step tutorial on F5 TTS - the free, locally voice cloning tool that'll blow your mind π€― on my "Thorsten-Voice" channel.
Check it out here π
https://t.co/YTwIVisgws
#TextToSpeech#AIVoice
π Thorsten-Voice turns 5! ποΈ
To celebrate, I'm gifting the community ALL voice datasets in their original 44kHz quality - now in one place on @huggingface!
Higher quality, easier access - same CC0 license π
Get it here: https://t.co/mC3eDioiay
#TTS#OpenSource#AI
Let's goo! F5-TTS π
> Trained on 100K hours of data
> Zero-shot voice cloning
> Speed control (based on total duration)
> Emotion based synthesis
> Long-form synthesis
> Supports code-switching
> Best part: CC-BY license (commercially permissive)π₯
Diffusion based architecture:
> Non-Autoregressive + Flow Matching with DiT
> Uses ConvNeXt to refine text representation, alignment
Synthesised: I was, like, talking to my friend, and sheβs all, um, excited about her, uh, trip to Europe, and Iβm just, like, so jealous, right? (Happy emotion)
The TTS scene is on fire! π
3 steps to run @huggingface "Parler TTS" AI Voice on your local machine. New tutorial video out now π!
My step-by-step technical tutorial is now available on my "Thorsten-Voice" youtube channel.
https://t.co/pNvaXF11J4
π£οΈ September 30th is #InternationalTranslationDay, and we want to celebrate the Language Leaders. They are helping us build an open-source, privacy-focused voice assistant that anyone can run.
Thanks to all your hard work helping us over the Year of Voice and beyond! π»ππ»
π₯NEW VIDEO: "State of free text to speech | 2024.08".
First look + audio samples of new cool #opensource#tts#ai projects and voices on my Thorsten-Voice #youtube#channel π₯³.
β‘οΈ https://t.co/AYjSPbm55x
hey so i now live in leipzig and am looking to connect with linguists or people working on NLP/ASR. or just other cool people. im so bored!! (i speak english, german, dutch, and spanish)
twitter do your thing please??