OT: Fun fact about austism...

[Anthony William Sloman]

Phoneticians divide the speech sounds we make into about 200-odd distinguishable phonemes. Specific languages use a subset of these, often lumping similar sounds together. English has about 44 phonemes

https://magoosh.com/english-speaking/44-phonemes-in-english-and-other-sound-blends/

though it varies a bit between dialects. All English speakers hear \"r\" and \"l\" as distinct phonemes. Japanese speakers have to make an effort to make the distinction.

When normal adults listen to an unfamiliar speaker, they adjust the threshold between phonemes to match what that speaker produces. It happens rapidly and automatically, so you can induce the threshold shift. and demonstrate that it has happen in a ten minute lab experiment. The shift is specific to the individual speaker.

Children don\'t do it, and it turns out that adults who been diagnosed as on on the autism spectrum don\'t do it either.

--
Bill Sloman, Sydney

 

[Carlos E.R.]

On 2022-05-02 10:37, Anthony William Sloman wrote:

Phoneticians divide the speech sounds we make into about 200-odd distinguishable phonemes. Specific languages use a subset of these, often lumping similar sounds together. English has about 44 phonemes

https://magoosh.com/english-speaking/44-phonemes-in-english-and-other-sound-blends/

though it varies a bit between dialects. All English speakers hear \"r\" and \"l\" as distinct phonemes. Japanese speakers have to make an effort to make the distinction.

When normal adults listen to an unfamiliar speaker, they adjust the threshold between phonemes to match what that speaker produces. It happens rapidly and automatically, so you can induce the threshold shift. and demonstrate that it has happen in a ten minute lab experiment. The shift is specific to the individual speaker.

Children don\'t do it, and it turns out that adults who been diagnosed as on on the autism spectrum don\'t do it either.

I am interested in this you say, on the last two paragraphs.

Do you have links that expand on this?

--
Cheers, Carlos.

 

[Martin Brown]

On 02/05/2022 09:37, Anthony William Sloman wrote:

Phoneticians divide the speech sounds we make into about 200-odd
distinguishable phonemes. Specific languages use a subset of these,
often lumping similar sounds together. English has about 44 phonemes

https://magoosh.com/english-speaking/44-phonemes-in-english-and-other-sound-blends/

though it varies a bit between dialects. All English speakers hear
\"r\" and \"l\" as distinct phonemes. Japanese speakers have to make an
effort to make the distinction.

Their \"l/r\" sound \"r\" is somewhere in between English l and r and cannot
occur without a trailing vowel (usually a short u in most cases). Only
the Japanese letter n can occur on its own without a trailing vowel.

This means that the correct answer in Japanese \"English\" to the question
what do you call an Apple(Ringo in Japanese) in English is: Apuru

The corresponding gotcha for the Western ear in Japanese is that the
length of a vowel sound really matters and we normally ignore that. Many
UK dialects would be incomprehensible if you didn\'t ignore vowel length.

eg.
Sco-ne vs Scon for example depending on where you are in the country.

My own name Martin relies on long vowels in Japanese Ma-chin.
(no R available and no Ti phoneme) Luckily it ends with an \"n\".

If you are listening for vowel length then that is what you hear!

--
Regards,
Martin Brown

 

[Anthony William Sloman]

On Monday, May 2, 2022 at 10:16:14 PM UTC+10, Carlos E.R. wrote:

On 2022-05-02 10:37, Anthony William Sloman wrote:
Phoneticians divide the speech sounds we make into about 200-odd distinguishable phonemes. Specific languages use a subset of these, often lumping similar sounds together. English has about 44 phonemes

https://magoosh.com/english-speaking/44-phonemes-in-english-and-other-sound-blends/

though it varies a bit between dialects. All English speakers hear \"r\" and \"l\" as distinct phonemes. Japanese speakers have to make an effort to make the distinction.

When normal adults listen to an unfamiliar speaker, they adjust the threshold between phonemes to match what that speaker produces. It happens rapidly and automatically, so you can induce the threshold shift. and demonstrate that it has happen in a ten minute lab experiment. The shift is specific to the individual speaker.

Children don\'t do it, and it turns out that adults who been diagnosed as on on the autism spectrum don\'t do it either.

I am interested in this you say, on the last two paragraphs.

Do you have links that expand on this?

Not yet. It should be published fairly soon.

--
Bill Sloman, Sydney

 

[Anthony William Sloman]

On Monday, May 2, 2022 at 10:26:02 PM UTC+10, Martin Brown wrote:

On 02/05/2022 09:37, Anthony William Sloman wrote:
Phoneticians divide the speech sounds we make into about 200-odd
distinguishable phonemes. Specific languages use a subset of these,
often lumping similar sounds together. English has about 44 phonemes

https://magoosh.com/english-speaking/44-phonemes-in-english-and-other-sound-blends/

though it varies a bit between dialects. All English speakers hear
\"r\" and \"l\" as distinct phonemes. Japanese speakers have to make an
effort to make the distinction.
Their \"l/r\" sound \"r\" is somewhere in between English l and r and cannot
occur without a trailing vowel (usually a short u in most cases). Only
the Japanese letter n can occur on its own without a trailing vowel.

It\'s a mora-timed language, as opposed to English, which is stress-timed and French, which is syllable-timed.

https://en.wikipedia.org/wiki/Mora_(linguistics)

This means that the correct answer in Japanese \"English\" to the question
what do you call an Apple(Ringo in Japanese) in English is: Apuru

The corresponding gotcha for the Western ear in Japanese is that the
length of a vowel sound really matters and we normally ignore that. Many
UK dialects would be incomprehensible if you didn\'t ignore vowel length.

eg. Sco-ne vs Scon for example depending on where you are in the country..

Dutch - on the other hand - definitely attaches phonetic signficance to vowel length. I never had any trouble registering that. whether I reproduce it as reliably as I should is an open question. My spoken Dutch seemed to be reasonably comprehensible, if vilely accented, so it was probably good enough.

My own name Martin relies on long vowels in Japanese Ma-chin.
(no R available and no Ti phoneme) Luckily it ends with an \"n\".

If you are listening for vowel length then that is what you hear!

Everybody listens for vowel length, but precisely what they extract from the information does depend on the language being spoken (or the language they think is being spoken. The famous joke is about the zookeeper who was told that there was a \"moose loose\" and made sure that the speaker wasn\'t Scottish before raising the alarm.)

--
Bill Sloman, Sydney