Hacker News Reader

Comment by WalterBright | Hacker News Reader

On a pragmatic note, the more of an accent a person has, the harder I need to work to understand them. Also, the more accent, the better quality I need from the phone to understand them.

Voice quality from worst to best:

1. cell phone

2. wired phone

3. Skype

4. in person

My hearing has declined somewhat (too much fooling about with hotrods), making the situation worse.

Another thing I've noticed when listening to a talk show. It's easier to understand the host than the guest, especially when the audio is speeded up. Apparently, a professional host enunciates words better.

I've looked around for a speech coach to improve my enunciation for the presentations I give, but all I could find were autism speech therapists and singing coaches, sigh. (I want to improve my enunciation because the easier it is for people to understand my speech, the more they'll pay attention to my presentation. I know that I sometimes give up listening to a presentation when I have to work too hard to understand the speech.)

I've been doing a lot of video conferencing for work these days (like everyone else) and decided to trash my $5 mike and get a semi pro one so people could hear me clearly.

View on HN

There is no such thing as "more" or "less" accent from a purely linguistic point of view. You can of course distinguish accents by how close they are to some standard accent, but English doesn't even have a single standard accent.

The idea that "some people have accents, but I speak the pure (or a purer version of the) language" plays right into the notion of accentism.

Tho German has standardized variants for Germany, Austria and Switzerland, regulated by the international body of of the Council for German Orthography.

Imho it's also kind of reductive to make "accentism" out as something purely discriminatory, when the matter of fact is that particularly a lot of regional accents are often unintelligible for people who are not familiar with them, even when they are native speakers of the language without ascent.

In Germany one can grow up and live in the middle of a state for their whole lives, and still heavily struggle with understanding the local accent of a smaller nearby town.

And while that is certainly an enrichment on the cultural diversity side to have that many, and so different, accents, it also makes communication more difficult, more prone to fail, thus ultimately more likely for misunderstandings to happen.

There are standardised variants of grammar and orthography, not necessarily of pronunciation. There is a more loose standard (never written down anywhere explicitly I believe, not surprisingly, since who's gonna read a phonetic treatise?) in Germany in terms of pronunciation, which is certainly what most people in the media, try to speak, but you won't have any such luck in Switzerland, neither with actual Swiss German, nor with the way Swiss people pronounce Standard German (it's interesting: I can usually roughly tell where a Swiss person is from when they speak German).

Not understanding, or having more trouble with, certain accents doesn't make you an "accentist", but thinking that this is a fault on the other person's side is.

> In Germany one can grow up and live in the middle of a state for their whole lives, and still heavily struggle with understanding the local accent of a smaller nearby town.

Does this really still happen nowadays?

> And while that is certainly an enrichment on the cultural diversity side to have that many, and so different, accents, it also makes communication more difficult, more prone to fail, thus ultimately more likely for misunderstandings to happen.

Sure, but I guess that's the price you pay for diversity, one I'm personally glad to pay.

"Does this really still happen nowadays?"

Not a direct answer to your question, but I live in the Netherlands which is a lot smaller than Germany, about 300x200 km. Every village has their own accent/dialect and every region often has their own words or even sub-language. I live 50km from Fryslan, but I cannot make any sense of Frysian language. The same for Limburg or Flanders when people have a deep accent.

Sometimes people with such a strong accent get on tv, and they provide subtitles, otherwise it is just incomprehensible to most Dutch people. It is often a culture of relating to someone, like "We are from the same background". At my chess club there are 3 persons that know Frysian, when they decide to talk Frysian, it is better to just walk away :)

Frisian isn't too relevant to a discussion of Dutch accents, because Frisian is universally considered to be a different language than Dutch. After all, Frisian and Dutch split off from their common ancestor already in the first millennium AD.

And I guess even the spelling of the language name is up to debate ?

The English Wikipedia article on Frisian doesn't even list "Frysian" as an alternative spelling, so this is an error on the part of the OP.

It may not list it as an alternative English spelling, but the page for the main dialect of Frisian does indicate that the language's own name for itself is spelled with a Y.

https://en.wikipedia.org/wiki/West_Frisian_language

> There are standardised variants of grammar and orthography, not necessarily of pronunciation.

It usually directly translates to pronunciation as that's a rather big part of spoken Hochdeutsch.

> Does this really still happen nowadays?

Yes, a lot. Thick Bavarian accents are often considered intelligible in the rest of Germany. Even inside Bavaria dialects can range from "easily understandable" like a lot of Franconian, to something so thick that even other Bavarian's might struggle with understanding, like some regional Upper Palatine dialects.

Often even small towns will have their very own accents.

> Sure, but I guess that's the price you pay for diversity, one I'm personally glad to pay.

I guess I'm just too utilitarian for that, my ideal version of humanity would speak one common language to improve cooperation and understanding among each other.

Dialects are cool and interesting, but there are already so many languages to learn, expecting people to learn a whole bunch of dialects for each of those, in addition to the language itself, is quite an impossible task for the vast majority of people who don't learn languages for a living.

> Yes, a lot. Thick Bavarian accents are often considered intelligible in the rest of Germany. Even inside Bavaria dialects can range from "easily understandable" like a lot of Franconian, to something so thick that even other Bavarian's might struggle with understanding, like some regional Upper Palatine dialects.

Somewhat related story about Swiss German, although it is more of a dialect than accent.

I self-studied German for two years - Duolingo, consuming a lot of German media, chatting via video with different speakers online. I thought I had a pretty good grasp on basic conversational skills.

When I finally went to Germany for the first time, I had a connecting flight in Zurich on Swiss Air. After boarding the flight I realized I could hardly understand anyone. Most of the staff and passengers were speaking Swiss German. I panicked a good part of the flight about my misplaced confidence in my fluency.

Luckily on the trip through Germany I had no problem with Hochdeutsch. However there definitely were very noticeable differences in accent and word usage as I moved from Schleswig-Holstein in the north to Bayern in the south via the east.

> It usually directly translates to pronunciation as that's a rather big part of spoken Hochdeutsch.

Not really, since there are lots of small pronunciation differences (e.g. the pronunciation of "ä", the quality of "i" in words like "Kirche", etc.) that are not distinguished in writing. At the same time, some "Bühnenhochdeutsch" pronunciations are not even captured by orthography, e.g. syllable-final -ig being pronounced as -ich or the voicing of the s sound before vowels at the syllable onset. In fact, especially with phenomena such as reduction of "-er" etc. in standard German, Swiss people will usually claim that they speak "German as it is written" even though that is wildly different from how most Germans speak.

Also, I mean, modern German and its orthography was modeled in large parts after Saxon, but nobody would think nowadays that Saxon constitutes "standard German" in terms of pronunciation.

> Often even small towns will have their very own accents.

But that wasn't your claim. You claimed that dialects are unintelligible from town to town which is usually very rare, since dialects tend to form a continuum.

> I guess I'm just too utilitarian for that, my ideal version of humanity would speak one common language to improve cooperation and understanding among each other.

That's not a linguistic question anymore, but one of personal values. However, it is worth noting that cases of language death often are accompanied by culture death as well, and deep feelings of shame and guilt within the respective communities.

Saarbrücken, the barely moderate-size city I studied at in Germany, has three distinct dialects. It lies right on the French border. At least one dialect has heavy French influence and one does not.

Most of the local students I talked to spoke German that was very close to Hochdeutsch plus some linguistic peculiarities and slang terms. As far as I could tell this was the dialect they used at home, when drunk, talking with childhood friends, etc., and I could understand it easily.

Every once in a while, though, I would meet someone (usually of a slightly older generation) whose dialect I found completely unintelligible.

I'm not sure how easily the average local German could understand the different dialects. Most of the local people I talked to regularly were computational linguist students, German teachers, or worked in the international student office, so they had a particular interest in learning languages.

> There is no such thing as "more" or "less" accent from a purely linguistic point of view.

Bit of a nitpick, but I think that's only true of so-called native accents. [0]

If a community speaks the language natively, in a certain distinctive way, that's called a native accent, and as you say it's meaningless to refer to any native accent as more or less accented than another. (edit I suppose in the extreme cases this might not be true, as a community's dialect might evolve into a distinct language, but that's by the by.)

With non-native accents though, the accent is the result of a limited linguistic ability. If you learn French, you'll have less of an English accent as you become more skilled, as the accent is the result of your native English leaking into your French pronunciation.

[0] https://en.wikipedia.org/wiki/Accent_(sociolinguistics)#Non-...

At this point, we're moving into meta-debates about the meaning of the word "accent". But the way most linguists understand the way "accent" is just as a pattern of pronunciation, so even with non-native speakers, they don't have "more" or "less" accent, they rather change their accent to a more native-like accent.

case in point: Standard German doesn't pronounce syllable-final "r", so someone who speaks English with a German accent will tend to speak with a non-rhotic accent, as most English accents. If they subsequently move to the US and adopt rhotacism, are they now having "more" or "less" accent?

I think the issue is explained quite nicely here: https://web.archive.org/web/20081008000501/http://linguist.e...

I've heard people from India say they don't have accent, but so do newyorkers. Being native Spanish speaking, I find easier to harder english accents:

- Russian, Italians

- English RP (BBC)

- "Educated" American English (politicians, etc.)

- rural accents from the US

- India

- Ireland

- Australia

- UK popular accents

> There is no such thing as "more" or "less" accent from a purely linguistic point of view.

Pretty sure what the parent meant by "the more of an accent a person has" was "the more a person's accent deviates from my own andthe majority of my community", but wanted to be more concise (if less precise).

I'm guessing you're right, in the context, though, it's not the best shorthand to choose. The unrecognized perception that one's own accent is not actually an accent is the keystone of accentism.

> The unrecognized perception that one's own accent is not actually an accent is the keystone of accentism.

Is there anything research to support that claim? Seems like it is just a regular tribal thing (us vs them, in-group vs out-group).

English is not my first language, so when I speak, people don’t always understand me. Sometimes I have to repeat words more slowly, or replace words with synonyms. This happens because I do not pronounce the words in the way people around me expect to hear.

The explanation above is somewhat long, so I would like to shorten this. I have been saying “I have an accent” instead.

Is “accent” the right word to use in this case? If not, which word is?

Accent is the correct word.

> I do not pronounce the words in the way people around me expect to hear.

This is an interesting way of putting it - an accent is a collection of linguistic tendencies that together can clue a listener in on your social class, geographical origin, or other biographical information. However, I think it is possible to make a distinction between "having an accent" and simply mispronouncing words. Stressing the wrong syllable is an example of the latter. Both having an accent and mispronouncing words can lead to misunderstandings, but one is easily correctible and the other is not.

"I have an accent" as a kind of shorthand for "I have a non-standard accent" is in theory fine, since most people know what you mean.

The reason I pointed it out, though, is because the reverse direction - "you have an accent, which is why I don't understand you" - especially when directed at actual native speakers implies (even if not intended) some sort of deficiency on the part of the interlocutor.

> There is no such thing as "more" or "less" accent from a purely linguistic point of view.

I assumed it was obvious that I meant "relative to my accent", which is what Europeans have noted as a "California accent".

You basically said "everyone has an accent", and this is true. Regarding understandable English though, some accents are MUCH harder to understand than others.

from your perspective

you can't just assume that everyone finds the same accents easy or difficult

My perspective is that, for most talks in English, the speaker would want to be understood by native speakers of American English.

I do assume (and believe correctly), that the further an accent strays from what you would hear on your standard American news channel, the more difficult it is for the average native speaker of American English to understand.

https://en.wikipedia.org/wiki/American_English

To your point about the difficulty of finding a good coach to train enunciation, you should look for a parallel track. A trite example is the scene from Pumping Iron where Schwarzenegger takes ballet classes to help improve his posing. You want someone who can train you in an area that requires the same skills but for a different purpose. I would look to stage acting where the performer is required to speak clearly and forcefully enough that a large audience can understand what they are saying while still maintaining control over their voice. Maybe a Shakespearean acting coach could help you.

Agree on all fronts. As a non-British person living in England, some local accents are nightmarishly different from the basic "international English". For example, I absolutely love Irish and Scottish people, their history, their culture, their traditional attitudes etc etc, but when they "go thick-accent" I can't fucking understand half of what they're saying. I feel like the Robin Williams sketch on landing in Scotland.

> decided to trash my $5 mike and get a semi pro one

Did the same, and also got decent headphones, and it made a ton of difference on both sides - people understand me better and I understand them better. Some people are so impressed they actually ask how I get to "sound so good".

I dread to answer my landline phone now, I keep it working just for emergencies.

I know that acting lessons often have enunciation training. That's partially why there is a "TV accent" that's more or less regarded as the standard US accent.

As a non-native English speaker, I've found that the best way to make myself understandable is to enunciate the consonance clearly. It sounds weird sometimes, and clearly not native, but it's always clear.

Another thing I've trained myself to do is the control of rhythm and stresses. I got this idea from a book on poetry, which classified English as a stress-timed language. Coming from a mother tone of Chinese, I'm not used to grouping rhythm centered on stress syllables. For me, stresses used to live just within an English word, as notated by the dictionary.

When I started to pay attention to a sentence-level flow of stresses, I finally understand how English speaking people gloss over some syllables and emphasize on others. I've even noticed that many eloquent speakers naturally slip into a iambic pentameter when they talk.

Regarding speech understanding: I think even in everyday situations, we rarely understand other people 100% and derive the missing bits via context. I watched a lot of movies in the English original with English subtitles over the past years, and because of that, I am now used to be able to understand movie conversations 100%. Whenever I watch a movie in my native language (without subtitles), I sometimes have to pause and replay a scene to fully understand the dialogue. Sometimes it is virtually impossible, as the actor just mumbles a few syllables. When I ask other people watching with me what the actor just said exactly, they usually also don't have a clue, but it doesn't seem to bother them. Usually, the answer is something like "he certainly meant A, because B just did C, and now he is excepting D".

This happens to me even with some American accents. Or I'm just losing my hearing :-/

I’ve been watching a lot of tech YouTube during the lockdown to play with new languages and the accent (and audio quality) tax on me as a listener is real. If I’m trying to learn something from you and it’s mentally hard to start with, if I have to burn extra cycles replaying what I heard to extract the words, I’m going to get a lot less out for the time and effort I put in.

It’s in some ways akin to trying to read content with constant pop-up ads obscuring the message and needing my attention to dismiss.

Yes, if I can't understand what someone is saying due to a thick accent, I will immediately close the window.

You might look into drama coaches that give elocution lessons. I imagine that the skills are fairly similar.

I used to do a bit of drama and choir and I found the warmup exercises to be helpful for enunciation and such. A brief warmup before a meeting might be a good idea too. The major downside is that it looks and sounds ridiculous, but it will limber up the muscles in your mouth and face.

Regarding voice clarity on smartphones, it’s quite terrifying to see the difference in audio quality (recorded or during calls) from one instant message app to another, on the same phone.

For example, from worse to best:

- wechat - WhatsApp - Threema - Voice memo.

How are other apps scoring at this? (Telegram / Skype / Naver / etc.)

> the more of an accent a person has, the harder I need to work to understand them.

Implicit in this statement is the (incorrect) assumption that you yourself don't have an accent. It makes no sense to complain about people having "a lot" of accent.

The quoted statement of fact does not imply that the poster believes himself to have no accent. The implication in this statement is the (plainly true) assumption that from the perspective of a listener, the greater difference a speaker has to their own dialect is what is always meant by 'more' of an accent.

Furthermore, the presence of several dialects does not preclude there being a 'central' dialect, with 'less' of an accent, being most easily understood by all- eg the columbus oh 'midland' accent being the traditional american 'radio standard' for its neutrality and strong general pronunciation.

Yes, exactly.

Sadly, cell phone voice quality has not improved since my first cell phone 30 years ago.

So I think the speaker and microphones probably were better in the older phones, at least as I recall them.

But "HD Voice"[1] as they call it here is really noticeable for me, and with earbuds or similar it's quite good. It's quite apparent when I'm out in the forest and it drops out due to poor signal level.

[1]: https://en.wikipedia.org/wiki/Wideband_audio

My relatively new Nokia feature phone has very good voice quality.

I can't imagine that the voice quality is the same, given things like VoLTE and WiFi calling basically turning cellphones into VOIP systems. I feel like this is one of those things where the improvements were so gradual that you don't recognize them happening. But, if you actually talked to someone on a 30 year old cell phone and network, you would maybe wonder that you ever thought that was clear.

I am surprised Skype ranks second. Can it be because it's usually used with headsets on a computer?

I am surprised that Skype is ranked better than a wired phone.

It has a higher bit rate.

A wired phone (POTS) doesn't have "bit rate", it has bandwidth. It could very well be that VOIP uses a lower bitrate than Skype, or Skype uses a better algorithm, but I would lump VOIP in the "Skype" category, not the "wired phone" category.

I do know that POTS is analog (at least the wire to my house is, I presume at some point it becomes digital), not digital, but the larger point stands.

A wire phone (POTS) definitely does have a "bit rate". The entire POTS back-end system is digital, and each circuit gets 8000 8-bit samples per second, or 64kbps, uncompressed.

Anything with a bandwidth has bitrate:

https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampli...

Well, to put it simply, no.

Nyquist's theories (and Shannon's sampling theorem based on it) only concerned themselves with the time-based discretization of analog signals. As Shannon showed, to accurately reconstruct a signal of bandwidth x, you need to take samples at a constant frequency of 2x.

But when speaking of digital signals, discretization is only half the story. The other part of the story is quantization, i.e. how many bits to use to represent each sample. And that amplitude quantization is a lossy operation, regardless of Shannon's theorem about discretization in the time domain. You can't simply say that an analog signal of 40kHz therefore has a bitrate of 80kbps -- using just one bit to represent each sample will not accurately reproduce the original analog waveform, unless the original waveform was a square wave (and you would still need that bit of information to accurately reconstruct it).

So what, then, is the bitrate of a discretized, analog signal? It is still infinite, because the amplitude of every sample can be any element of R! Only once you decide on a quantization resolution, i.e. once you determine your acceptable level of signal loss, can you speak of "the bitrate of an analog signal".

But that bitrate is only an approximation of the original signal, Shannon notwithstanding.

Your flat dismissals do nothing to disprove Shannon's math. Come back when you have equations.

I am not so sure: according to Wikipedia, Skype uses SILK, which has bitrate of 6-60 kbit/s [0]. In contrast, G.711 (often used for classic landline telephones) has a bitrate of 64 kbit/s [1].

I can only speak from personal experience: my cheap standard landline has much better audio quality and lower delay times than any Skype call I have made.

[0] https://en.wikipedia.org/wiki/SILK

[1] https://en.wikipedia.org/wiki/G.711

Is this bitrate comparison an "apples to apples" comparison?

I know when talking about audio compression in the frame of music reproduction AAC 256Kb/s is similarly performant to MP3 320Kb/s - and I think there are better performing audio codecs still.

Bitrate only tells a little bit of the story. There are high pass and low pass filters, audio compression, data compression, latency, error correction and a number of other things at play with these codecs to hit a desired bitrate with the goal of maximizing for highest legibility over unreliable connections. This leads to a number of compromises; in the case of Skype this means you can have huge variance in quality, and landlines are consistent but incredibly limited in reproducible bandwidth and dynamic range.

This is no criticism, interesting comment. But...

It's really fascinating that the HN crowd always obsessively needs to put a technical spin on any single topic posted, whatsoever.

I'm an engineer, I can't help but solve problems. For example, while the Iraq war was going on and soldiers were getting killed by IEDs, I kept designing a vehicle in my head to be resistant to such infernal devices. The armored vehicle designers likely know how to do this far better than me, but I couldn't help it.

But maybe not. I bought a Generac generator the other day, and the controls on it are so horribly designed I want to scream. It isn't horrible because they saved money, it is pointlessly, stupidly bad, and Generac has been making generators forever. It baffles me.

If it is not “critical” then why did you choose to use the word “obsessive”, instead of a more neutral term?

I mean it's literally a board that caters to engineers and technology enthusiasts.

Reply on news.ycombinator.com