Synthetic Language Puns

It seems that there's a tendency for IE languages to become more analytic. My language Greek, for example, has lost (compared to ancient Greek): 1 case, 2 moods, 1 voice, 1 number, the infinitive, and has much simpler noun declensions; and at the same time has developed many new periphrastic features. Then, of course, we all know English (compared to old English).

So are there any IE languages that are outliers to this trend?

Strategy ape here.

I’m deep into Part 2 of The Sun Never Sets on Citadel (HOLY SHIT IT’S GETTING WILD), and in light of dlauer’s awesome post today I’d like to clear up some of the terms we use.

I see many apes throwing around the terms “synthetic” shares or “rehypothecated” shares interchangably. They are NOT interchangable - they do NOT all mean the same thing. Here is what they mean:

#SYNTHETIC A synthetic share (or any synthetic position) is using a couple of options to mirror the price of a stock.

• HOW: 1) You buy sell a put. 2) You buy a call. 3) You do 1+2 in a way that they “mirror” the price of the stock. Meaning: a $1 gain on the stock = $1 gain in your position, a $1 loss on the stock = $1 loss on your position.
• WHO: Hedge Fonds and other market players who want to adopt a position but don’t want to own the stocks. They don’t care about voting, and maybe they don’t want to declare a position or affect the price of the stock with their purchase, or whatever. Also the options position is generally cheaper.
• Additional features: you can pair it with a bond, if you’re really into cosplaying your options to make them look extra stock-like.
• Do Apes have these?: NO.
• SO... this is like MEGA BLOCKS. Nobody believes these are Legos, except people who know nothing about Legos.

#PLEDGED A share that a Market Maker says it will get from the market. The DTCC says a pledged share is real, so it’s real.

• HOW: 1) A MM takes an order for a share, and fills the order even if it has no share yet. 2) The exchange and DTCC want you go have confidence in buying, so they back up that pledge and say that it’s a legitimate share. They are the final say. It’s a share.
• WHO: Market Makers make these. This is a legitimate practice, nothing shady – if done correctly (I’ll explain in a minute). But they CAN be abused. MMs are known to “rent out” these shares to Hedge Fonds (*cough* Citadel and Melvin *cough*) at the expense of investors.
• Additional features: can make a real share disappear IF the MM buys a share off the market and closes out the pledge (basically tells you it's a real share, then takes one off the market to even it out). OR becomes an FTD if the MM is late in closing the pledge.
• Do Apes have these?: YES
• Are these naked shorts?: NO – but if the MMs neglect them they will be! (will always be a real share to Apes tho)
• SO… Your mom cal

An Apple research team explores multiple architectures and training procedures to develop a novel multi-speaker and multi-lingual neural TTS system. The study combines speech from 30 speakers from 15 locales in 8 languages, and demonstrates that for the vast majority of voices, such multi-lingual and multi-speaker models can yield better quality than single speaker models.

Here is a quick read: Apple Neural TTS System Study: Combining Speakers of Multiple Languages to Improve Synthetic Voice Quality.

The paper Combining Speakers of Multiple Languages to Improve Quality of Neural Voices is on arXiv.

Just wanna know how this sub feels about introducing synthetic languages (like Esperanto) under socialism to improve international cooperation and/or for communication among different national groups in multinational countries.

Also, are there any theories as to why many languages evolve to become more analytic?

I hear a lot about synthetic languages when every people talk about languages they find interesting, which I'm not complaining about okay because they can be really interesting. Especially how many things they can do, like languages with noun cases can have free word order, or how poly-synthetic languages can make an entire sentence into one long word.

I don't however hear much about analytic languages. Probably because this sub is full of primarily native English speakers and English is fairly analytic itself so analytic language features probably go overlooked. But still, I was wondering, are there any language features that analytic languages don't that either synthetic languages don't have or are much rarer among them?

Hello, my name is Filip. I'm the founder of GenHub (https://genhub.co/) and I made a programming language for designing genetic circuits.
I am looking for synthetic biologists that will use my language in their research to prove it's viability. I'm looking for someone that actively uses genetic circuits in their experiments.
After the experiments are proven successful, we can extend our cooperation through a contract that benefits both of us. The terms can be discussed privately.

So I want to know a bit more about premodern and especially non-european linguistic tradition. It seems to me that in older european linguistic tradition, synthetic languages were regarded as more refined, this was probably due to the classical image of Latin and Greek. For non-european language, syntheticness was regarded as quality for their refinement, IIRC this was remarked for Nahuatl for example, reasoning it to be a classical language aswell. On the other hand languages with less synthetic morphology were decribed of having less complex grammar, isolating languages having no grammar. You've probably heard this statement about chinese too.

So what I'm asking is the reverse, how did grammarians from China (or better, members of the scholastic tradition of the sinosphere), before the introduction of modern (western) linguistics describe and regard more synthetic languages (Korean, Japanese, Mongolian) in contrast to the isolating structure of Chinese and others like Vietnamese? Was there a reverse stereotype. How did they describe the morphology of synthetic languages in their own terms?

I'm trying to wrap my head around the idea of analytical languages vs synthetic languages. I get the idea in theory, a synthetic language has a high morpheme-to-word ratio so you get very specific words with a lot of syllables and an analytical language has a low ratio and you need lots of small words to convey a specific meaning. The issue I have is when these are put into use. How do we know that languages we consider synthetic (Finnish, Inuktitut, etc.) aren't actually using a bunch of small words in a phrase instead of a bunch of syllables in a single word? So far as I can tell, there doesn't seem to be anything phonological that distinguishes between two monosyllable words vs one bisyllable word.

I could have a synthetic word from some arbitrary language (I've made one up for this example) that is nadukemet where naduk is a root word for 'person' and the suffix -emet is means 'a group of'. So, nadukemet in this hypothetical synthetic language means 'a group of people'; how would I know the language isn't synthetic and I'm really hearing naduk emet. The meaning is still the same but I'm using two different words.

Hi. I'm an ethnic Circassian. As far as I know, over the past 10 years, I'm the leading producer of content and materials to learn the Circassian language.

I'm interested in creating a text-to-speech (TTS) engine / synthetic voice that can do a reasonable job with the Circassian language.

I've played around a fair amount with Amazon's Polly voice services, and I've built a pretty decent tech stack that allows me to build a variety of language assets for supported languages.

Obviously, Circassian is not a supported language...

Can anyone here help me understand whether / how I can build a synthetic voice for Circassian?

I have high-quality audio of 1,500 unique phrases, recorded by a professional, native speaker of the language that can be used to train an engine. I also have all of these phrases spelled out, and getting them converted into IPA shouldn't be too big of a deal.

Any suggestions would be greatly appreciated. TIA.

A strong focus on correlations between theoretical linguistics and NLP. https://youtu.be/dMo_eLNsg08 | https://twitch.tv/wasspen - a new channel - cause sharing is caring 😊

I was curious because most Indo-european languages are losing things like dual number, declensions, etc. If the general trend is for languages to lose its declensions, how did PIE become so synthetic in the first place? Sorry if this is a dumb question, I'm a noob here.

English: Almost immediately shortens "coronavirus" to "corona"; later decides it's too vague and tries to switch to "covid-19". Also can't help but smash some words together, so people with inadequate reactions get called "covidiots". I'm pretty sure I'm missing some other neologisms, but these are pretty characteristic.
Russian: Also tries "корона" but the word is already being used for "a crown". After some hesitation it stumbles upon the tansliteration "ковид", deems it worthy to be a noun and a stem, and in just a few weeks there are 90k google results for the adjective "ковидный" (...кашель, госпиталь, пациент, etc), including official speeches (!). Meanwhile folks derive another adjective "ковиднутый" which might mean either a covidiot or someone who's ill and the speaker feels strongly negative about it.

Perfect example of how languages work in general :)

Can anyone suggest a scenario under which a modern synthetic Indo-European language (Spanish, French, Russian etc) would gradually lose its synthetic features and become a tonal analytical language? What processes could drive such a change and how, grammatically and phonetically, would this change happen? Has anything like that ever happened in real life? This is a thought experiment of sorts, so any plausible scenario should work.

Provided that the language is not related to the learner's native language, and all other factors (orthography, phonology, loanwords etc.) controlled? For example, is Malay (an analytic language) inherently easier to learn for a Japanese native speaker than Latin (a synthetic language) is? My intuition says yes, but I would like to see if there is any research on this topic.

By becoming more analytic, I basically mean two things: loss of case and use of articles.

To illustrate my point, here are two maps:

[Articles in languages in and around Europe](https://en.wikipedia.org/wiki/Article_(grammar)#/media/File:EuropeArticleLanguages.png)

Number of grammatical cases in the European languages

As we can see from the maps, there is the divide between the Analytic West and Synthetic East. This begs the question: where and when did this trend of analyticness start? Can it be explained by intensive language contact among Western European countries in the Middle Ages? Can this also be the reason why the Baltic and Slavic languages did not follow this trend?

Of course, there are exceptions in both families (Icelandic and German for Germanic, Romanian for Romance--these languages have retained case). Iceland and Romania are separated geographically from other members of their families, but how has German retained case?

PHONETICS

Dumaket follows a very loosely defined phonetic system. As a result, different speakers may use similar but different phonemes (for example, š could be pronounced as either /ʃ/ or /ʂ/). The consonants are pretty straight forward; the only thing that might need specifying is j, a palatal approximation if on its own. If j occurs after a consonant, the pronunciation can be up to interpretation (for example, mj could be pronounced as /mj/, /mʲ/, or /mi/).

The vowels are somewhere around these sounds:

i /i,ɪ/
u /u,ʊ/
o /o,ou/
a /ɑ,ə,ɔ/
e /e~ɛ/

Words and affixes ending in a vowel will have an implied consonant (written in parenthesis when the word is presented in a dictionary). This consonant is written or said when a suffix placed after it begins with a vowel. For example, "gjewu(k) (to eat) has a k sound in its verbal root in "gjewukek" (I eat) but not in "gjewulawatek" (I habitually eat).

The implied consonants only work within words; it does not split between words. You cannot say "lokwas akisek" (using the words "lokwa(s)", "bear" and "aki(s)", "to look at"); you can say "lokwa akisek" (I look at the bear). In situations like this where you have to vowels on either side of the word boundary, you use a glottal stop to separate the words.

When two stop consonants of the same articulation (e.g. g and k) are affixed together, they are separated with -a-.

VERBS

Verbs are a foundation in Dumaket; there are many nouns with verbal roots ("godešapet", "bed" from "godeš", "to sleep") and a single verb can form an entire sentence ("walanžukagodešapetekayetuk", "I hug him/her/it in bed").

Dumaket verbs do not require a subject to make a complete sentence. In the case of many spontaneous and natural events, the verb may simply be used on its own without the use of a dummy subject. For example, "vaduk" (to snow) can be used on its own to say "it snows".

Intransitive verbs can be used in two ways: using a noun or using a suffix. "Mjela godeš" (the deer sleeps) demonstrates the use of a noun, "mjela" (deer). Because Dumaket lacks pronouns that can be used with verbs, suffixes are used instead, as demonstrated by "godešek" (I sleep).

Transitive verbs are a bit more complicated:

If the agent and patient of the verb are both nouns, Dumaket uses an SOV structure: "kowul mjela wolka" (the wolf kills the deer).

If the agent is denoted by a suffix and the patient is a noun, Dumaket uses an OV structure: "mjela wolkasuk"

Hi everyone,

Just curious if anyone here has heard about the fairly recently developed programming language for synthetic biology (CRN++) developed by researchers at University of Texas?

If so, what are your thoughts?

• Will the industry take it up?
• Does it seem intuitive / easy to learn?

Would love to hear more!

I have a system on paper that allows biologists to design genetic regulatory networks using traditional programming syntax, but I don't know exactly how to make it a reality. Overall, the author would describe which transcription factors trigger which genes and the program, having an extensive library of known proteins and the DNA that codes for them, would output a string of nucleotides. I want to open a text editor, type something like:

if (protein_1_concentration > .2){
     transcribe(protein_2);
    }

and the output would be something like:

 [ATATCGGG][ATS][TCGATGGCCGA][TAG]
 [Protein 1 promotor sequence][Start codon][gene for protein 2][stop codon]

Would I need to design a brand new language or could I simply make a library for python. Any suggestions from programmers and biologists alike would be greatly appreciated. Thank you!

EDIT, Hopefully this is a more comprehensive design document:

An 'if' statement would trigger the creation of an 'operon'. An operon is a sequence of nucleotides consisting of a promoter, a start codon, all desired genes, and a stop codon in this form:

[PROMOTER][START][GENES][STOP]

The condition following the 'if' denotes the promoter sequence. for example, lets say the promoter sequence for TRANSCRIPTION_FACTOR_A is ATAGCGA. This:

if (transcription_factor_a){
...
}

would be this:

[ATAGCGA][START][GENE][STOP]

In all organisms, the start codon is ATG. There are several stop codons, but we'll use TAG. Start and stop codons are required during translation (RNA -> protein) so the ribosome knows where the code for a single protein begins and ends.

The command within the curly parentheses denotes which genes will be transcribed. Let's imagine the gene for PROTEIN_A is TCGAAATGG. This:

if(transcription_factor_A){
transcribe(protein_A);
}

would be this:

[ATAGCGA][ATG][TCGAAATGG][TAG]
or 
ATAGCGAATGTCGAAATGGTAG

Protein_A could be a transcription factor itself, creating the potential for incredibly complex networks of gene regulation.

a single promoter could be responsible for translating multiple genes as long as there are start and stop codons surrounding each gene.

The transcribe function would have to include the names and nucleotide sequence of various proteins and transcription factors.

Does anybody know of any languages which have a full personal paradigm of synthetic passive forms? I.e., instead of "I was seen", "he was seen", "they were seen", etc, one that has verbal inflections to indicate the personal object.

All the languages I'm familiar with, even synthetic ones with comparatively complex verbal paradigms, revert to indicating personal objects of passives with a standalone pronoun instead of a personal suffix.

An Apple research team explores multiple architectures and training procedures to develop a novel multi-speaker and multi-lingual neural TTS system. The study combines speech from 30 speakers from 15 locales in 8 languages, and demonstrates that for the vast majority of voices, such multi-lingual and multi-speaker models can yield better quality than single speaker models.

Here is a quick read: Apple Neural TTS System Study: Combining Speakers of Multiple Languages to Improve Synthetic Voice Quality.

The paper Combining Speakers of Multiple Languages to Improve Quality of Neural Voices is on arXiv.

An Apple research team explores multiple architectures and training procedures to develop a novel multi-speaker and multi-lingual neural TTS system. The study combines speech from 30 speakers from 15 locales in 8 languages, and demonstrates that for the vast majority of voices, such multi-lingual and multi-speaker models can yield better quality than single speaker models.

Here is a quick read: Apple Neural TTS System Study: Combining Speakers of Multiple Languages to Improve Synthetic Voice Quality.

The paper Combining Speakers of Multiple Languages to Improve Quality of Neural Voices is on arXiv.

A strong focus on correlations between theoretical linguistics and NLP.YouTube |Twitch Link - a new channel - cause sharing is caring 😊

Is Hebrew an analytic or synthetic language?