Help! :-( confused about testing

So I've made my own minimal pairs cards with Anki.from day one I encountered the predictability issue discussed on another thread, and installed the randomise add-on ( thanks for that!) And I'm really confused! I understand the idea is to go in blind, don't learn anything else, just start contrasting sounds. But what if I simply don't know the difference? It doesn't matter how many times i hear the word, if its foreign to my brain then i don't know how to match what I'm hearing with what I'm seeing. It's like, if someone shows you the colour red and says is this red or blue? It doesnt matter how many times they ask and how many times you look- if you dont know what red is, you can never answer. I have all these cards piling up in my review day after day because hearing the word means nothing to me and i think 'well it could be either! How to tell?' What do i do? I am thinking i have to jump ahead and learn IPA and pronunciation rules etc to create a context, just to go back and actually complete the pairs test. I'm happy to but it totalky contradicts the very specific deliberately laid out step by step method and I'm trying to perfectly adhere to the program. I get upset when i don't understand then it's not fun and exciting anymore :( any feedback really appreciated! Really :)

* Originally posted by Meher Baba.

Didn't find what you were looking for?

New post
Was this post helpful?
0 out of 0 found this helpful



  • Hi Meher,

    I understand your frustration. It can be very difficult to train your ear for the new sounds in a new language. I've been learning Mandarin for a while now and I still sometimes have problems distinguishing between tones used in that language. Gabes pronunciation trainer has been a wonderful resource for helping me with that.

    Getting your ear trained first seems absolutely essential to me for all the reasons that Gabe laid out in his book. As he said, getting instant feedback on whether you have chosen the correct option is absolutely essential.

    One thing that has really helped me is that after I have made my choice and seen whether I've gotten it right or wrong I replay the sound in Anki a few times while I am looking at the spelling of the word. I find the combination of visual and audible feedback is a huge help to cementing it in my mind. Having done that a few times really increases the chances that I will be correct the next time that card comes around.

    Hope this helps.

    * Originally posted by Hacksaw.
  • @Hacksaw is right on the money about the benefits of replaying a few times. However, with minimal pairs cards, I like to use the option of playing both choices as the answer – this is an easy modification in Anki, as per one of the videos.

    But another aid that I tried for minimal pairs, seemed to work well, even though I didn't think it would. Even if I had no idea about which of the two choices was more likely, if I tried to mimic the audio, I was able to get the correct answer much more frequently ... it just seemed that the process of getting the tongue in the right spot the reproduce the sound helped a lot.

    * Originally posted by Lance.
  • Good tip Lance! Nice to leverage muscle memory to help your recall.

    Regarding playing both sounds, are you having Anki automatically play them both when it shows the back of the card or are you just making both sounds playable if you hit the play button?

    * Originally posted by Hacksaw.
  • I just play them both in order, with of course the correct one first. That is, I change the last line of the card from:
    {{Recording 1}} to {{Recording 1}}{{Recording 2}}
    {{Recording 2}} to {{Recording 2}}{{Recording 1}}

    I haven't tried to do anything more sophisticated than this. Not sure I'd be able to do so!

    With the Mandarin Trainer, I've got good value from the Min Pairs, but haven't been able to stay the course with the the main part, as I'm still struggling with tone discrimination, at least via that mechanism.

    I seem to have got more value from tone testing on sites such as and the tone pairs recordings on YoYo Chinese and Sinosplice. But it is a struggle without access to native speakers. Need to get onto iTalki, I think!

    * Originally posted by Lance.
  • Hi Lance,

    I have noticed a couple of the tone cards in the Mandarin Trainer have the wrong tones in them. I verified with my Mandarin tutor and submitted bug reports. Not sure if they got fixed or not.

    Plus the normal Mandarin tone changing rules such as what happens with two consecutive 3rd tones being spoken as a 2nd and a 3rd can make it a bit tricky. I just go with what I hear and still count myself as correct if I said 2nd and 3rd when it was really 3rd and 3rd.

    Also since tones are based on the speakers vocal range they can sound different depending on who is saying them. I have a lot easier time hearing the tones from some people as opposed to others. But then I've always been tone deaf when it came to music so it's no surprise I have similar problems with Mandarin.

    Amusing anecdote: Often when I ask my tutor (who is a native Mandarin speaker) what tone the word she just used was she actually has to stop and go through the tones one by one until she finds the right one. It comes so naturally to her she never has to think about it. That's what I'm striving for, but I've got a loooooong way to go.

    * Originally posted by Hacksaw.
  • Hi Hacksaw and Lance. I am really grateful for your response. So some feedback for the benefit of others:
    1. Having your encouragement and shared experience made an immediate difference in my perspective. So, anyone feeling discouraged - don't do it in isolation!
    2. From your response I quickly found peace by appreciating the sheer fact of repetition - that is the solution. It simply takes time. We are pushing the boundaries of our audible comfort zone and that's a challenge. And a 'Respect!' for your language (Mandarin) being significantly harder than mine!

    Thanks for suggestion about mouthing the sounds Lance. Excellent idea. However, this runs counter to the idea of sticking JUST with ear training before moving to the next step

    I'd be curious to know what Gabe feels about this, if you're reading Gabe, as I keep encountering this issue - conflict between doing steps sequentially or mixing them up for trouble shooting. The latter is intuitive but Gabes counsel is the former.

    My constructive feedback for future editions of the book and website is to demarcate the learning steps more succintly. For me with my OCD brain trying to totally map everything out in total clarity to create a sense of comfort - each step in the methodology blurs and melts into each other a bit too much. I appreciate with the sheer complexity of the subject its hard to make natural boundaries. I am immensely grateful for the gift we've been given here

    I'm not using an official pronunciation trainer so perhaps I've got to forge my own path a bit little more

    My minimal pairs are as confusing as 'rock and lock' apparently are for Japanese. Another tip : I found a youtube of someone pronouncing each of the sounds in my target alphabet. I converted it into an mp3 file, and used Audacity to single out the identical sounds for paired comparison, and saved them separately.
    So I now have an audio file with just those sounds in isolation as pairs. I am listening to them while I follow their description on a piece of paper.
    Actually why not put that into the trainer? Good thinking :) It correlates with what Gabe said the studies say about feedback being far superior to sheer repetition.

    The predictable playback order issue in Anki - which can be fixed with a randomise ad-on - is all explained here:

    Gabe suggested a Fluent Forever app to replace Anki in the future - that would rock!
    Programmers send in your emails :)

    * Originally posted by Meher Baba.
  • PS - another idea. I've taken those isolated sound pairs and viewed the wavelength in Audacity. To try and get a sense of how they differ.
    This could be the most distinctive and definite visual cue we can get ? Incoprorated into cards, even? It must be an image the brain can really relate to

    I understand we are essentially trying to create new neural pathways here

    * Originally posted by Meher Baba.
  • Hi Mehar,

    I'm glad we could help. Hang in there! It definitely does get easier over time.

    DEFINITELY make an audio trainer for yourself. Without the feedback you are definitely hobbling yourself. I'm lucky enough to have an Android phone so I take my pronunciation trainer and such with me everywhere. If I've got a few extra minutes of downtime I can get some practice in.

    That's an interesting idea about viewing the audio wave in Audacity. I'm not sure if it would be all that helpful for me right now, but it is certainly interesting enough to take a look. Thanks!

    * Originally posted by Hacksaw.

Please sign in to leave a comment.