#tts — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #tts, aggregated by home.social.
-
Was just thinking. We've got a good choice of TTS voices on macOS and iOS now, some available and some in TestFlight betas.
Would love to see the True Voice voices there too. I have them for NVDA on Windows.
Can someone make that happen?
-
#TTS Lector on F-Droid, bare-assed document text to speech. THANK GOD when I can find a no-nonsense absolutely basic toof for my phone nowadays. I just wanted to listen to epubs/PDFs like I used to, in the crappy clearly-machine voice, no AI voices weirdly inhaling, no data mining colonoscopy. Less than a MB. GPL3. What a respite from the travails of a simple utility search. This is why I use F-Droid.
https://f-droid.org/en/packages/io.github.quantember.lector/
UPDT: I've ended up using this by copying several minutes worth of text and pasting it into the app. This works very smoothly for html copied straight from gutenberg's website. For epub and pdf it was reading every carriage return as a paragraph pause. Copied and pasted from html it pauses only with actual paragraphs or periods.
-
Hey, Pa? How come we got no good critters this year?
-
Does anyone our there know a good open source text-to-speech software?
I'm looking for something that I can upload a document, like an article, and have it give me an audio file that I can export, not just read it back to me.
-
Beginning my search for non-ai TTS website readers, the only one I found so far was on fdroid- it's called eSpeak (which is a Really cool older project- totally synthesized speech), but it's hard for me to understand or listen to for long.
I guess I want the type where it's actual human voice recordings of each syllable, then reassembled. Sorry I forgot what it's called.
TIA!#TTS #nonAi #DisabilityAccess #braindamage #neurodivergence #anxiety #selfmedication #Audhd #audiobooks
Part 2 of 2 -
Voicebox 공식(https://voicebox.sh)
GitHub jamiepine/voicebox(https://github.com/jamiepine/voicebox)
-
So I really wonder if there is any, / are several, hubs for finding new piper voices? Sure, the lib in the app or nvda addon has a few, sometimes peple here on mastodon train 1 or 2, but I'd like to find more unique ones more easily rather than sticking to what the library has and maybe 1 or 2 external models. So if anyone knows websites for it, do let me know. Looking for more natural stuff, rather than tts replicas or low quality models if possible.
#a11y #piper #tts -
CW: USSR
So according to Reddit, the Soviets had something called the Fonemophone-4 speech synthesizer, based on the Elektronika-60 computer, developed between 1978 and 1991. C'mon, #nvda and #tts guys. Where's my Fonemophone-4 NVDA addon? All you'd have to do is emulate an Elektronika-60; hardware emulation is already being done for Dectalk and Speech Proes and Keynote Gold, so this is a solved problem, right? LOL JK. #USSR #retrocomputing #nvdasr -
Мы опубликовали стабильный, быстрый, качественный и доступный синтез ещё для 29 языков России
В прошлый раз у нас получилось покрыть 20 самых популярных языков России и СНГ с рядом оговорок. Но основным недостатком моделей было то, что значительная группа языков была вообще не покрыта, в основном многочисленные языки Кавказа, Дагестана и Чечни. С лингвистической точки зрения из популярных крупных языковых групп были не покрыты абхазо-адыгские и нахско-дагестанские языки. Что сделали в этот раз?
https://habr.com/ru/articles/1065252/
#silero #синтез_речи #tts #texttospeech #нейросети #озвучка #языки_россии #тюркские_языки #кавказские_языки #нахскодагестанские_языки
-
(now with updated audio at a better pitch)
Woo! I got the DECtalk on ESP32 board working! Here's a long sample of its output (standard USA Paul voice, slow 200 wpm rate), reading from Chapter 4 of "Augustus Carp, Esq."
Hardware cost for this is $30-40 or so. It's supposed to be compatible with JAWS.
Project URL:
https://dectalk.homerow.net/ -
No joy with the ESP32-C6 based DECtalk clone. Checked wiring. Tried two different audio DACs. Fiddled frantically with the volume control. No output
The thing does respond to serial commands and seems to be behaving in that way. But a very frustrating day.
-
#HashtagGames
Try doing the #CarTroubleASongOrPoemas a TEXT-to-SPEECH thing here:
User: [email protected]
p/w: demopass
(or create your own)doesn't anyone find it useful, or HILARIOUS?
the #HashtagGames is honestly what i thought of when I was making it interactive.
-
Released my first Godot Asset! A retro voice synthesizer :mycomputer:
Asset: https://store.godotengine.org/asset/eibriel/tts-079/
Try it on Itchio: https://eibriel.itch.io/scp-079-voice
-
@Extelec this might be old news for your #RC2040 project, but there's a CP/M translation of the CTS256A text to allophone chip that might allow speech conversion on-device:
https://github.com/GmEsoft/TRS80_CTS256A-AL2 -
Text to speech stopped being a novelty when the vocoder got fast.
The pipeline is three stages: text analysis decides how words should be spoken, a model predicts a mel spectrogram, then a vocoder turns that into a waveform. Early vocoders took seconds to render one second of audio. The current ones stream, which is the only reason voice agents can hold a conversation.
Where the tools actually stand: https://www.aitools9.com/ai-text-to-speech/
-
👀 Convertire un EPUB in audiolibro sul tup PC: un container e un comando. Kokoro TTS lo fa capitolo per capitolo, tutto offline. 👇
https://gomoot.com/kokoro-tts-come-far-leggere-pdf-epub-e-articoli-al-tuo-pc/
#epub #kokoro #pdf #tts
