Post a sample of your pronunciation here!

[kvetch]

Hey, thought I'd post a sample here.

The text I'm reading is the first paragraph of this

http://mandaread.com/read/1573/1/hongbao-cu#

The site has pinyin.

 

我 今 年 跟 一 个 中 国 家 庭 过 年 。 这 次 过 年 跟 我 以 前 过 年 很 不 一 样 。 礼 仪 很 多 ， 饭 菜 很 多 ， 也 可 以 了 解 到 家 人 之 间 的 复 杂 关 系 。 我 今 年 也 有 机 会 更 深 刻 地 体 会 到 红 包 文 化 。

 

What I'd like to know is what words I'm pronouncing wrong( cheng is too nasally or I don't pronounce my h's right etc.)

Whether I'm stressing the right words or forgetting to stress anything. And if my tones are all over the place/ not there at all-I already realised the last character has wrong tones and I that I pronounced shen as sheng.

 

Thanks!

 

Btw does anyone else feel way too self concious reading aloud in their thinly walled dorms? I keep imagining the fuwuyuan laughing at me.

[陳德聰]

Where's your recording?

[kvetch]

Aw, hell.

Sorry, here it is.

Voice 275.mp3

[陳德聰]

Hi kvetch!

Sorry I have taken so long to take a look at your pronunciation sample, I listened the first time just before bed and fell asleep without finishing.

But! I have noticed a few things:

 

Your transition from tone to tone happens prematurely sometimes. This is not a huge issue, but it sounds a bit odd. Sometimes it is more obvious and sometimes it is more blurry. For this reason I had to open up the audio file in Praat to look at formants to reassure myself my ears weren't playing tricks on me.

What it should have been:

Wǒ jīnnián gēn yíge Zhōngguó jiātíng guònián, zhècì guònián gēn wǒ yǐqián guònián hěn bù yíyàng: lǐyí hěn duō, fàncài hěn duō, yě kěyǐ liǎojiě dào jiārén zhījiān de fùzá guānxì. Wǒ jīnnián yě yǒu jīhuì gèng shēnkè de tǐhuì dào hóngbāo wénhuà.

 

What your pinyin would look like:

Wǒ jìnnián gèn yíge Zhòngguó jìatíng guònián, zhècì guònián gèn wǒ yǐqiàn guònián hěn bù yíyàng: lǐyī hěn duō, fàncài hěn duō, yě kěyǐ liǎojiě dào jiàren zhìjiàn de fùzā guānxì. Wǒ jìnnián yě yǒu jīhuì gēng shènké de tǐhuì dào hǒngbào wénhuà.

I have bolded all of the tone mistakes, red denotes every time you changed a 1st tone into a 4th tone, blue is for every time you did something I'm going to call levelling out, and then the remaining mistakes I think are just from general misspeaking.

Tone issues:
What I would guess is that for the red ones, you are transitioning early. You are able to produce duō in isolation no problem, but it looks like when the target tone level of the following word is in the lower register, you are pre-emptively going down to get ready for that tone. Notice you have no problem with guānxì and jīhuì.

For the blue, I think you are recognizing that there are two tone level targets for these words, but for some reason you're not producing both. In lǐyí, it looks like you're just going straight to the end result of the tone rising. In fùzá, either you're doing the same thing, or you're pre-empting the 1st tone in guānxì that follows it. For "gèng shēnkè", I think it's possible you did a tonal swap, or you were pre-empting the 1st tone from shēn. The reason I have grouped hóngbāo changing to hǒngbào into the blue one is that you didn't produce a rising tone there, and even though you didn't fall on hóng, it still sounds like a 3rd tone if you leave it trapped in the lower register.

Pronunciation issues:
For the most part, your pronunciation is good. It sounds a bit nasal overall, but I am not very sensitive to nasality in general so that's difficult for me to comment on.

One thing that struck me is your "e" vowel as in "yíge" and "shēnkè". It sounds like [ɔ], which would almost change the pinyin to "yígo" and "shēnkò".

[andyfastow]

Hi everyone, I'm new to these forums and also new at studying Chinese. So far I've mostly been using Pimsleur audio lessons (just about finished the first set of 30 lessons) and I would love to get some feedback on what little I can say so far.

 

In the audio file I say 6 phrases from the Pimsleur Mandarin I course. I haven't looked up the pinyin for every word, and in fact just went through the pinyin section of FSI for the first time a few days ago. I'm mostly just trying to mimic the Pimsleur recordings as this is apparently the point of their method.

 

I'm feeling like Pimsleur isn't getting me very far and I hate that there is no written material included with their audio so I'm thinking of spending the bulk of my time on FSI and private tutors going forward instead of Pimsleur.

 

Anyway, I'd love some feedback on how this sounds. I get enough quizzical looks from native speakers that I've tried to say even simple things to that I'm already past being offended by any harsh criticism

 

I wrote what I'm trying to say in English below the cut. I'd like to know how much is intelligible without seeing it written down first. Any other feedback is of course welcome as well. Thanks!

 

Hello. I speak English. I speak a little Mandarin, but I don't speak well.

 

Excuse me, may I ask, where is the Beijing Restaurant?

 

My wife is American.

 

I don't understand what you say.

 

You speak too quickly.

 

Do you want to drink some beer?

phrasesfrompm1.mp3

[roddy]

I'd say that's mostly comprehensible, but you've picked up a few mispronunciations which are getting in the way. Most prominently (for me, at least):

 

huì in wǒ huì shuō (I [can] speak) is missing the h sound

xǐang in the last one is missing it's 'a' sound

 

Maybe go back to the audio and compare those with what you're saying?

[andyfastow]

Thank you Roddy, I will do that.

[roddy]

You might also want to look at Audacity and this recent post from Tamu (afternoon and a flask of tea required). That'll let you put your pronunciation and Pimsleur's on a loop so you can get forensic on listening to the difference. 

[陳德聰]

Is your voice creaking because you're deliberately speaking in a lower register than you're used to?

I'm impressed that Pimsleur can get one this far, but I am biased against it for reasons I won't mention here haha.

 

When you say "I can speak English", your "speak" has a vowel in it that is not quite right. This could cause misunderstandings if you're saying things to people out of the blue.

The pinyin is shuo, but your [ɔ] sound is coming out like [a]. So your "you speak too fast" kind of sounds like "you brush too happily"

 

Then, when you ask if I'd like to drink some beer, you should be saying "ni xiang he yidianr pijiu ma?" But instead you are saying "xing". Notice the lack of [a] vowel there, and then you left off a syllable at the end.

 

Like roddy said, you are comprehensible, though there is much room for improvement. If your few mistakes here are indicative of systematic issues, then you could be saying "wa" or "-ua" every time you're supposed to be saying "wo" or "-uo", and "-ing" every time you're supposed to be saying "-iang". These can cause a lot of confusion if your context is not as obvious as it is for these few phrases.

Using audacity is a fantastic tool for working on your tones, and for the time being perhaps re-examine your vowels and that creaky thing you're doing with your voice.

[kvetch]

陳德聰, Thank you, that was a really detailed and useful review. Never knew I was screwing up the easiest tone. I have a tendency to speak too fast and slur my words together, so now my main goals are to a) go over my tones in praat  b)  speak slower.

[andyfastow]

Thank you for your reply 陳德聰, I appreciate your feedback. I'm going to keep repeating the FSI module on pinyin and pronounciation every other day or so and I'll pay particular attention to the vowels.

 

I think the mistakes I made in those recordings are indeed mistakes I make all the time. I'm going to do what Roddy suggested about recording audio files to see if I can correct some of these mistakes.

 

In those files I was speaking normally, but I have a poor quality headset and/or sound card. I have "microphone volume" and "microphone boost" as high as possible and checked "auto gain control" for added volume, but my recordings are still barely audible. I actually boosted the volume of those clips a little using Goldwave and then filtered out a mean hissing noise in the background which might account for my voice sounding creaky.

 

I think I'm going to check out some electronic stores in Taipei this weekend and see if I can find a half decent USB microphone that won't break the bank. I want to start using my microphone for italki tutoring, LE on Skype, and doing what Roddy suggested. My current setup probably isn't going to cut it.

 

Thanks again for your suggestions!

[歐博思]

08年度中国最具幸福感10大城市出炉

2008年12月26日在云南昆明，揭晓了2008年度中国最具幸福感城市，其中10座城市入选，并举行了隆重的颁奖仪式。这10座城市分别是杭州 宁波 昆明 天津 唐山 佛山绍兴 长春 无锡 长沙。其中杭州由于连续五年在调查推选活动中名列前茅而获得金奖。

此次评选由新华社《瞭望东方周刊》联合中国市长协会从今年四月开始，调查内容涉及自然环境，交通状况，发展速度，文明程度，赚钱机会，医疗水平，教育水平，房价，人情味，治安状况，就业机会，生活便利共12项指标，采取专业公司调查与公共调查相结合的方式进行，近300家媒体参与，共计700 张调查问卷和 7000万次网络投票。

 

Please give me a good roasting 

 

 

*edit* : deleted file

[thefamousamos93]

 

08年度中国最具幸福感10大城市出炉

2008年12月26日在云南昆明，揭晓了2008年度中国最具幸福感城市，其中10座城市入选，并举行了隆重的颁奖仪式。这10座城市分别是杭州 宁波 昆明 天津 唐山 佛山绍兴 长春 无锡 长沙。其中杭州由于连续五年在调查推选活动中名列前茅而获得金奖。

此次评选由新华社《瞭望东方周刊》联合中国市长协会从今年四月开始，调查内容涉及自然环境，交通状况，发展速度，文明程度，赚钱机会，医疗水平，教育水平，房价，人情味，治安状况，就业机会，生活便利共12项指标，采取专业公司调查与公共调查相结合的方式进行，近300家媒体参与，共计700 张调查问卷和 7000万次网络投票。

 

Here's my shot at it  

 

140312-052731.mp3(mp3).mp3

[陳德聰]

@欧博思
Okay so... I really don't like this excerpt as it tests your character recognition more than it tests your pronunciation, and a lot of the issues seem to stem from either not actually knowing what the character is or from slow recall of character leading to delayed awareness of word boundaries. I think it would be infinitely more useful to just speak casually for a few minutes using as many different words as you can, and then transcribing afterward. This way you are using only words you actually know, and it leaves just the pronunciation to judge.
 

 

08年度中国最具幸福感10大₁城市出炉

2008年12月26日在云南昆明，揭晓了2008年度中国最具幸福感城市，其中10座城市入选，并举行了隆重的颁奖仪式。这10座城市分别是杭州、宁波、昆明、天津、唐山、佛山、绍兴、长春、无锡、长沙。其中杭州由于连续五年在调查推选活动中名列前茅而获得金奖。

此次₂评选由新华社₃《瞭望东方周刊₄》联合中国₅市长协会从今年四月开始，调查内容涉及自₆然环境，交通状况，发展速度，文明程度，赚钱机会，医疗水平，教育水平，房价，人情味，治安状况，就业机会，生活便利共12项指标，采取₇专业公司调查与公共调查相结合的方式进行，近300家媒体参与，共计700(万???)张调查问卷和 7000万次网络投票。

As a general note I find that sometimes your 2nd tones sound a bit unnatural, despite the contour being correct, their starting point seems independent of the prior words' ending points. I marked them in blue. They don't impede meaning, they're just a bit odd.

Also, is your first language German or another one that doesn't have many [dʒ] sounds?? I noticed that sometimes your /zh/ sounds like /ch/ and some of your fricatives are a bit muddled. Examples would be what I numbered as (2), (6), and (7). Your 此次 was not cǐcì, but ǐ[s/z]ì. Your 自然 came out as sìrán, and your 采取 sound like cǎicü.

 

Other issues involved your pacing and rhythm, which I think is a matter of improper stress. (1) was odd, not tonally but just in terms of how you stressed 10 but almost omitted 大. Your 揭晓 was odd, but I couldn't figure out why so I didn't include it. I think this also does sometimes affect your tones though, because of how you pronounced 投票 at the end with a strong 4th tone on 投 and essentially 轻声 on 票.

 

You also have a bit of a strange vowel quality sometimes, like (3) and (5) where 华社 sounded almost like huáshà and 中国 sounded almost like 中guá.

I also can't figure out why (4) sounds odd to me, because tonally it's fine, but it just... That one character felt so disjointed. No impedement to meaning, just a bit jarring.

The red are tones that are either weird or wrong:

 

颁，分，新，今，3，& 7 were times that you sounded like you were pronouncing these words in 4th tone. All were followed by a tone that has a lower starting pitch, so I assume this was because of premature tone transition.

选 came out sounding like 2nd tone.

文明 was pronounced so fast that all I heard was wēnmīng.

医疗 came out sounding like yīliǎo.

人情味 was weird to me and even though I looked through Praat for a while on this word I couldn't figure out exactly what was so weird. The 人 felt more tonally exaggerated than the 情, and then 味 started lower than I expected it to.

治 came out as a first tone instead of fourth.

标 seemed kind of high, and had a little bit of a rise that I think came at the beginning from your transition after 指. It sounded a bit like a super high second tone, but there was nowhere to rise since you were already at the top of your register.

You made 合 into a 轻声 and slurred it together with the following 的 which sounded unbalanced with the weight of the first character in the word 结合. You did a similar thing in 7000万次 with 次. It's a little bit less obvious with 次 since it's a measureword.

参与 you neglected to pronounce the tones for. This is similar I think to 文明 in that you were focused on different words and just sort of slided through the word on the same tone. 参与 sounded like two really low first tones in a row.

You inserted a 万 after 700, unless there were actually 7 million questionnaires handed out. In which case I don't know why the original text didn't have this. Either way, your stress on the 万 turned the following 张 into "zhǎng" instead of the 1st tone it was meant to be.

I think that is all the points I had meant to discuss when I marked up this passage I'm sorry it has taken so long for me to get to this, I was reluctant due to the content of the passage and also the amount of schoolwork I am supposed to be doing (though this has become suitable procrastination material).

In summary:
- You have a tendency to transition early from 1st tone into the following tones, creating what sounds like a 4th tone.
- You have a tendency to neutralize tones where it may not be appropriate, sometimes creating an illusion of being a different tone.
- Your second tones are sometimes too exaggerated/high.
- Your zh/ch, z/s, and q may need extra work. I think this might be an aspiration issue but I can't be sure from the data.

[davoosh]

I decided to give the piece from the previous post a go as well since 陳德聰 is giving great advice/criticism.

 

(08年度中國最具幸福感10大城市出爐

2008年12月26日在雲南昆明，揭曉了2008年度中國最具幸福感城市，其中10座城市入選，並舉行了隆重的頒獎儀式。這10座城市分別是杭州 寧波 昆明 天津 唐山 佛山紹興 長春 無錫 長沙。其中杭州由於連續五年在調查推選活動中名列前茅而獲得金獎。

此次評選由新華社《瞭望東方週刊》聯合中國市長協會從今年四月開始，調查內容涉及自然環境，交通狀況，發展速度，文明程度，賺錢機會，醫療水平，教育水平，房價，人情味，治安狀況，就業機會，生活便利共12項指標，採取專業公司調查與公共調查相結合的方式進行，近300家媒體參與，共計700 張調查問卷和 7000萬次網絡投票。)

 

It's been a while since I've spoken Chinese naturally, so I won't be surprised if this sounds odd/stilted (or just incorrect, haha).

 

I also have a question regarding tones and mispronunciation.

 

Let's say a foreigner sees a character s/he knows the pronunciation of, but forgets the tone. When s/he pronounces it, s/he doesn't stick to a clear 'citation' tone, but rather pronounces something strange and in between tones - would a native Chinese speaker mentally try to categorise this 'strange' foreigner's tone into one of the 4 'real' tones (since each of the 4 tones can have slightly different realisations), or would it just be heard as 'weird'?

 

Similarly, if a native speaker sees a character s/he thinks they know the pronunciation of, but not tone, would they just completely guess the tone (using one of the 4 'actual' tones)?

 

 

Thanks!

 

08年.wma

[歐博思]

Many thanks for the critique. I had to crank the mic boost on the computer, so that could have affected some of the 'data'  . Also, I have made a more conscious effort to emphasis 2nd tones recently; momma bear - papa bear - baby bear is coming!

 

What concerns me the most is your opening remark about word boundaries. I went over a few times to iron out any unknown characters before recording - so I was wondering if the phrasing problem of which you speak is more at the phrasal level/sentence level, or the word level? I feel I have a good grasp about word level grouping, but maybe not sentence level (to which you referenced above) ((and if I'm wrong about word level then it's also good to know!))

 

Take your time, I look forward to helping you procrastinate again later 

[陳德聰]

People people people please stop using this excerpt it's soooooooooo dry, just record yourself saying random words or having a conversation with yourself it makes life so much easier!

If you absolutely must do this excerpt, you can listen to the recording that goes along with it here.

===
 

@欧博思
I just went back and listened to it again to see. I think the issue might have to do with which words you're stressing in a phrase/sentence. I mistakenly brought up boundaries, but I don't think it has anything to do with those anymore. I think you're perhaps assigning stress to the wrong words, so certain other words don't get pronounced as much or for as long as they should, which results in a somewhat unnatural sounding flow.

 

I originally planned to go through the whole thing, but I lost steam after listening to this excerpt sooooooo many times haha.

[08年度][中国][最具幸福感][10大城市][出炉]

Here, it sounds like you separated the words into correct phrase groups, but when I went back again I noticed that 最具幸 has no evidence of downstep or reset whatsoever in your speech. You pronounced all three words not only with the same approximate length, but also the exact same pitch contours. This made it sound odd. If you made 最具幸福感 its own phrase, 具 as the verb should get the least amount of stress, whereas 幸福感 as the object should get the most. In this case, you should see the starting points of your 4th tones change throughout the phrase, with 最 starting at ~5, 具 starting at ~3.5/4, and then 幸 jumping up to ~5.5/6.

 

Then, when you said 十大城市出炉, you actually gave the most time to 城市. But you should have given 十大 and 出炉 almost equal weight, with 城市 being the least emphasized word.

2008年12月26日在雲南昆明，揭晓了2008年度最具幸福感城市

This went well. You stressed the right part of 最具幸福感 but there should be a bit more oomph in the 最 as well. Listening this time I noticed that 揭晓了's 晓了 seems under-pronounced. You can't change 晓了 into two non-existent tones since even 轻声 should have soooome sort of pitch.

[其中][10座城市][入選]，[並][舉行了][隆重的][頒獎儀式]
You can have a slight pause after 其中, and then again here, 十座 should get more emphasis than 城市. We already know you're talking about 城市, the focused portion is 十座. 入选 should get about the same amount of time as 城市. But in your version, you gave 十座 barely any air time, and then dragged out 城市入选 as if it was its own 成语. As for the second half, your 颁奖仪式 emphasizes 仪式 more than 颁奖, but again, the more important/new/"of interest" part of the phrase is 颁奖. Your 颁 gets lost here, where it should actually be kind of prominent.

I don't think I'm gonna go back through the whole thing, but I guess my suggestion would be to practice how you emphasize things. Usually, focused words are things that either implies existing alternatives (as in, not a 告别仪式, but a 颁奖仪式 <-- 告别 and 颁奖 would be the foci here), or is simply new information. In Mandarin the way to focus things is by increasing one or all of amplitude, duration, and range of pitch change. So play around with those notions, and pay attention to what people do to emphasize different words in a sentence, since your level is clearly quite high already, it's time to bring a more natural feeling to your speech~

===

@davoosh
To answer your first question, it depends on what the native speakers expectations of the foreigner's Mandarin are. If the native speaker assumes the foreigner's Mandarin is going to be crap, then when the foreigner speaks, the native speaker will usually pay less attention to the tones altogether, assuming they are all going to come out more or less weird. On the other hand, if the native speaker has a high opinion or expectation of the foreigner's Mandarin ability, then when the foreigner produces a strange tone that doesn't quite fit, whichever tone the sound is acoustically closest to will probably be what the native speaker hears. This can actually make it sometimes harder for intermediate learners to communicate things than beginners, since the native speaker assumes you mean exactly what you say and doesn't try to accommodate as much.

 

For your second question, I would say if the native speaker knows the "pronunciation" of the character, they definitely also know the tone.

And your critique is coming I promise!!!!!!

[MeiMay]

davoosh, I can't listen to the recordings but I'd like to expand on 陳德聰's reply to your second question. There is no way a native speaker would know the "pronunciation" of a character without knowing the tone as the tone is part of the pronunciation. It's like asking "if a native English speaker knows the final of a word but forgets the initial, do they just guess at the initial out of all the possible initials in the English language?" The idea of a native English speaker seeing a picture of a cat and remembering it's a word with "at" but having to guess if it was "cat", "rat" or "mat" or even "wat" is about the same as the idea of a native Chinese speaker knowing the consonant and the vowels but not the tone of a word.

[davoosh]

Thanks, 

 

 

I was referring more to the case of very rare/archaic characters, which the speaker might be able to guess the syllable based on other similar characters, but the tone may be different.

 

An example might be something like 尰 (for lack of better example), where a speaker might guess it to be zhong4. In these cases, would a native just guess with the tone of the 'similar' character?

[陳德聰]

My first guess for that character was chóng, but knowing that it is not zhòng, if it isn't chóng then I have literally no idea. I think the majority of characters are not 形声字, so it's not common that you can "guess" a character in the first place. After looking it up I see that it is zhǒng, and I feel confirmed in my feeling that to guess the character's pronunciation was a waste of time.

 

I just forced three of my friends to also guess that character's pronunciation, and all three of them just said "I don't know". At least that was what they said the first 7-10 times. Most said either chóng or zhòng based on the probability that it could be a 形声字, since if it was, those were the most likely pronunciations. But failing those, there is nothing left to guess... Made very clear when we all started saying things like "dà! xiǎo! pēn!"