#speech synthesis

[stevebattle]

A.G. Bear from Axlon (1985), by Ron Milner, Applied Design Laboratories, Grass Valley, CA. "A.G.” for "Almost Grown" Bear, was an instant success with its "growl-back" voice box (pictured), "great for the imagination. It's saying whatever a kid imagines it says." – Toy industry analyst Paul Valentine. "Nolan Bushnell requested a bear with a built-in recorder, to parrot a child's words. I suggested creating something with a personality. The result was A.G. Bear. … Axlon gave us two tough requirements for the bear's voice: "The voice box has to cost less than $2.50 to build," they said. Then they added, "… and we need one for the International Toy Fair – in two months!" That tight delivery schedule left little time for brain-storming. But we felt we'd have a more appealing toy if we could turn the talk-back concept into "growl-back." Our idea was to make the bear take the speaker's voice pattern, print its own inflection on it, then mumble it back. The result would be a soft murmur that sounded like a response to the speaker. We built a prototype of our concept on speculation, figuring that the best way to explain it was to show the idea in action. One week later, we demonstrated it to Axlon and they signed a contract. Over the next few weeks, we designed 3 different breadboards using discrete components. As our "mumbling" algorithms improved, each new version sounded better. The third time around we hit paydirt, meeting Axlon's cost and sound quality goals. Elapsed time? One month.” – Ron Milner

#cybernetics #speech synthesis #1985

[linguisticalities]

This is actually kind of cool. The Old English (only one I can attest to, except maybe Latin) sounds pretty convincing...

#AI #ancient languages #speech synthesis

[taperwolf]

Bell Labs, back in the 1960s, put out a number of educational book/kits about different aspects of signal processing and communications. One of those was this, "Speech Synthesis: An Experiment in Electronic Speech Production". YouTuber ParadisoModular got one from a science teacher back in the day, built it, and did some musical experiments with it back then, but he recently found an unused kit online and built it into a synthesizer module.

This circuit basically simulates the hunan vowel-producing apparatus through what's called "formant filtering". By shaping the lips, mouth, and throat, people create chambers that resonate at particular frequencies; the electronic kit does the same by using inductors and capacitors to create a series of three bandpass filters. For example, the 'ah' /a/ sound for the average male voice has its first formant at 730 Hz, its second at 1090 Hz, and third at 2440 Hz: you select these frequencies by swapping capacitors. The kit pushes a pulse wave into these filters to simulate the buzz of the vocal cords, and so you get the various vowel sounds out of it. ParadisoModular used switching chips to change the caps, so he can select multiple frequencies on the fly. It's a beautiful build.

Scans of the book are available online; I am going to have to build something similar for myself, though I'll probably move it to active filters rather than winding big coils.

#vintage electronics #electronic music #modular synth #homebrew electronics #speech synthesis #Youtube

[imlxh-music]

POCKET CALCULATOR

https://www.lexaloffle.com/bbs/?tid=49927

#pico-8 #kraftwerk #speech synthesis

[xlounzrhot]

I wanted to test out this voice synthesis program and 🤖😳I kinda like the way Arnold sound so no I can't stop thinking about a flirty, terrible, gamer clown who can't speed bridge to save his life😅

#xlounzrhot #elevenlabs #ai #speech synthesis #arnold

[medniex]

It's Jamuary! Month of raw music doodles.

#Jamuary2023 #jam #music #improvisation #instrumental #electronic music #Medniex #Alesis micron #Korg Kaoss DJ #Kaossilator 2 #KO2 #game boy #Chiptune #LSDJ #speech synthesis #vocoder

[vviederholung]

A demonstration of H.K. Dunn's and L.O. Schott's analog speech synthesis machine for Bell Telephone Laboratories, dating to 1951. According to the magazine article, available for download here, it was intended to study how humans make speech:

"Voice sounds were analyzed into their frequency components with a series of bypass filters... Then the frequencies which appear in a given sound were picked out--usually from a bank of tone generators--and mixed in the correct proportions to produce sound."

This audio clip contains a rendition of "Daisy, Daisy", that rather famous computer-sung song.

#analog #technology #speech synthesis #1950s #vintage #print #audio #speech synth always reminds me of florian schneider now

[richdadpoor]

Meta Releases SeamlessM4T Translation AI for Text and Speech

Meta took a step towards a universal language translator on Tuesday with the release of its new Seamless M4T AI model, which the company says can quickly and efficiently understand language from speech or text in up to 100 languages and generate translation in either mode of communication. Multiple tech companies have released similar advanced AI translation models in recent months. In a blog…

View On WordPress

#Applications of artificial intelligence #Computational linguistics #Creative Commons #Gizmodo #Human Interest #Internet #Large language model #Machine translation #Massively Multilingual #Massively Multilingual Speech system #META #Meta AI #Multilingualism #Paco #Paco Guzmán #Speech recognition #Speech synthesis #Technology #Translation

[cringywhitedragon]

Wanted to do some headcannons for some versions of the Microsoft Speech Synths (And a few others) from a personal Universe of mine

A lot of these are heavily inspired by the various MS Sam reads funny errors series I enjoyed watching on YouTube long ago and my own verse in a way so I give credit to the various creators whom I have forgotten the names of.

I’ve also used Picrew to give you guys a visualization on what my verse version of these guys look like (I was originally going to go with stick figures with these guys early on but decided to update their designs)

NOTE: I DO NOT CLAIM OWNERSHIP OF THE PICREWS USED, THEY BELONG TO THEIR RESPECTIVE OWNERS AND WILL BE CREDITED AND LINKED BACK TO

I’ll make a part 2 to this later on

A brief overview: All of these characters mentioned are what we would call a Synth. Synthians are a race that are almost indistinguishable from humans save for a few key features such as their voice and eyes. They function almost identically to humans and other living beings but also have their own quirks and goofiness that I won’t be going into detail on this post (And also because the personal AU they come from is a mish-mosh of different universes and frequently says screw the rules of space and time)

Synths can be divided into two distinctive categories: Regular Synths (IE: Sam, Mike, Mary) and Vocal Synths (IE: Miku, Len, Luka) based on their vocal patterns

Samuel Andrew Microsoft (MS Sam)

(Picrew by krmr, https://picrew.me/ja/image_maker/523501)

The obnoxious and chaotic of one the Microsoft triplets

Can be a bit of a jerkass at times (yet unlike some of his other counterparts, this version is actually a pretty decent Synthian) but does actually care a lot for his close friends/family and will stand up for them should the need arise

Has a speech impediment which is what helped him gain fame/infamy? (It depends on who you ask)

Enjoys giving people he’s friends with a hard time. His favorite target is his brother Mike since he finds it amusing when he gets riled up.

After him and his siblings were “retired” (aka fired) from their job as computer assistances and went their own separate ways, Sam spent quite a while as a drifter looking for a purpose. He took various odd jobs here and there up until he got to where he is now and reunited with his siblings aboard the starship they are now members of the crew (Once again, more personal lore stuff but Sam is the First Officer)

Is strangely obsessed with tacos (Something else I snagged)

Rather skilled at using vehicles and weapons

Does not have a filter and will actively curse, unless it is around younger Synths/humans/other races or if it is a professional matter.

Does not like Vista (Another obvious thing I snagged). This also goes for a specific drink known as Vista Soda which is another running gag I have with him.

Micheal Oliver Microsoft (MS Mike)

(Picrew by マサキ, https://picrew.me/ja/image_maker/54346)

The smart and serious one of the the Microsoft triplets

Often the butt-end of a lot of his brother’s jokes since he’s known to get snappy and have a pitch shift when he gets annoyed/scared

Currently works as a doctor aboard the Starship Whistler (one of the very few Synthian docs in the universe since most of their kind tends to work in jobs relating to either entertainment or technology) and has gained fame for his research into various diseases of the universe and helping to develop cures for them

A very minor one but he wears glasses

Has a tendency to speak very literally and sarcastically. Also fond of using big words but will often slip into tendencies that his brother has (usually cursing) when off duty or alone in his office.

Utterly terrified Hachunes and Mikudayios (Common organisms on what many Synths consider to be the Synthian home-world (aka “Earth 2”) outside of Earth. They are kinda like pigeons or something similar (even though Earth 2 does have pigeons), many Regular Synthians do consider them pests though the can be tamed/befriended and are just as intelligent as a human.) As soon as he sees or hears one of those creatures he will either bolt or lock himself in his office/room. Same goes for the song Lleven Polka which is often used as a measure to attract wild Hachunes, Sam has used this as a few times.

Can get serious when the time calls for it. This usually occurs during medical emergencies.

Has an obsession with soda.

Almost always has some sort of supplies on hand at all times. This especially goes for if the need ever arises to defend himself. Mike is very knowledgeable about most sentient races anatomy and can easily incapacitate an attacker (or if they are lucky, a simple sedative jab to the neck)

Marybell Anne “Mary” Microsoft

(Picrew by マサキ, https://picrew.me/en/image_maker/167775)

The motherly and kind one of the trio

Mary is a lot calmer compared to her brothers. She’s not as easy to set off and will not pick on others

Prefers to stay away from Sam’s antics and is usually the one to scold him when Mike isn’t around.

She’s a really good cook and went to culinary school sometime after her and her sibling’s retirement.

Opened a diner back on Earth 2 sometime before she joined the Whistler, it is currently being managed by a good friends

Very protective of her recipes (She will smack you with a pan if you dare try to steal them)

It’s very rare that you will hear her curse at anyone, unless she is really upset

Fond of reading and also enjoys gardening

Grandpa SAM / Samuel Mack PlainTalk (S.A.M / Software Automatic Mouth)

Grandfather to the triplets, and the one who raised the trio

Personality-wise, he’s best described as an older version of MS Sam who has mellowed out.

A big dork and somewhat of a flirt

While he may be an older Synth who mainly likes to play poker, he’s still rather found of video games (Something that he shares with his grandkids)

Good friend with a few Vocal Synths (Namely Leon, Lola, Miriam, and Meiko whom he met before the latter hit it big in a small bar on Earth 2 (The former two who now run the place), and even introduced Sam to them)

While he usually just goes by Grandpa Sam, his full name is a nod to his other family. He’s considered the father of the Mac Family of Synths (Something that he keeps well hidden from his grandkids.)

Annabelle “Anna” Blackcomb (MS Anna)

Not related by blood to the Microsoft Trio.

Anna is described as rather reserved and a bit shy

Not a lot do her past is known other then the fact that she was the replacement for the Microsoft trio and was later quietly fired

She has a stutter (speech delay)

She’s quite skilled with a sword

The Microsoft Trio never met her up until they joined the Whistler and found her on an ice moon (This has some major lore stuff for a character I’m not ready to fully reveal)

The person she would consider that she is closest to outside of an old pin pal from years ago is Mike. He was the one who found her and the one who spent the most time trying to get her to open up and trust them

#reshi rambles #vocal synth #speech synthesis #Microsoft sam #character headcanons #vocaloid #Microsoft Mike #Microsoft mary #Microsoft anna #fannon #microsoft windows #computers #Technology #headcannons

[pizzaronipasta]

#Meme-Repository #Moonbase Alpha #speech synthesis #aeiou #actually i have some new ideas what to use this for

[stevebattle]

"Pedro" the Voder – Voice Operation Demonstrator, by Homer Dudley and Helen Harper (1938), Bell Laboratories, Murray Hill, New Jersey. Another machine that Wowed the crowds at the 1939-40 New York World's Fair was Pedro the Voder, an early electronic speech synthesiser, invented by Homer Dudley. It could produce "20 or so different electric buzzes and chirps, which the operator would manipulate using 10 keys, a wrist plate and a pedal." It takes a skilled operator to elicit intelligible speech from the Voder, played as though it were an organ or piano. The pre-eminent Voder artist, Helen Harper, deserves equal credit for bringing it to life. Harper devised a year-long training programme for women to play the Voder, but only twenty made the grade. The Voder's nickname comes from the Brazilian emperor Dom Pedro, who on first seeing the telephone at the Centennial Exposition of 1876, exclaimed "It talks!"

#cybernetics #speech synthesis #1938

[miscpav]

DECtalk Flexidisc Demo 1984

#dectalk #speech synthesis #text to speech #flexi disc #dec #eva-tone #soundsheet #video #fran blanche #1984 #digital equipment corporation #flexidisc #ephemera #youtube

[witlessjester]

#Morrowind #Dagoth Ur #speech synthesis #catgirl #kajiit #a grand and intoxicating Innocence #Youtube

[ladygavgav]

Proofreading at Speed

New blog post: Proofreading at Speed.

One of the best ways to proofread a new piece is to leave it aside for a while, and then revisit it in the future. Writing and reading are two distinct processes, much like cooking and eating, My preferred formula is to leave one minute per word, or 24 hours, whichever is longer. To find this, divide the number of words in a piece by 60. So a 600-word short story would be set aside for 24 hours,…

View On WordPress

#blogging #editing #fiction #poetry #proofreading #prose #speech synthesis #writing

[aefensteorrra]

hmmmmm I miss my old life

#having to take a work break to cry in bed because I miss my ex? more likely than you’d think #feels extra shitty of me because I *am* dating someone rn #who knows about the future etc etc #but these feelings do not inspire me to read about neural speech synthesis or decipher my supervisor’s code

[richdadpoor]

ElevenLabs' AI Voice Generator Can Fake Voices in 30 Languages

What’s become one of the internet’s go-to companies for creating realistic enough visual deepfakes now has the ability to clone your voice and force it to speak in a growing variety of tongues. ElevenLabs announced Tuesday its new voice cloning now supports 22 more languages than it did previously, including Ukrainian, Korean, Swedish, Arabic, and more. ChatGPT’s Creator Buddies Up to Congress |…

View On WordPress

#Applications of artificial intelligence #artificial intelligence #Computer graphics #Culture #Deep learning #Draft:NoiseGPT #ElevenLabs #Entertainment #Gizmodo #Inflection AI #Internet #Mati Staniszewski #Mustafa Suleyman #PALANTIR #Paradox Interactive #Rutger Hauer #Speech synthesis #Technology