Barely progressing with listening comprehension

Open Forum
42

I just had a thought. Reading is painfully slow when I’m having to sound out words to figure out what they are. Until I am recognizing the entire words as units, it is so slow. But the ones that go fast are the ones where I recognize the whole word and don’t think about its letters.

Could there be something similar with listening, where we recognize phrases more than individual words?

I just went and listened to a bit of one of Alberto’s Italiano Automatico videos to try to see if I thought so (although Italian is not a language I practice or listen to podcasts in), and … I wasn’t impressed with this hypothesis of mine so much.

Some words work together and I think I understand them as a unit, like “al giorno”, but a lot of what he is saying, I think I am processing words individually. Like when he says “e audio disponible”, well, it’s a phrase, but I’m hearing the words individually.

I guess this just means it is actually not analogous to reading, where we really do skip processing the letters at all, and instead process the whole word (I believe). So now I don’t believe that hypothesis after all.

On the other hand, Alberto talks quite slowly, and pronounces the words kind of spaced out separately, so maybe this wasn’t the best test. But I probably couldn’t understand him if he were speaking at a more conversational speed.

1 Like
43

Might be a dumb question, but… are you listening to the AI speaker or selecting material where there’s a real speaker and you’ve somehow put it into Lingq?

1 Like
44

No dumb questions here! Always choose a native speaker. The point is for you to be able to understand real spoken language, so it needs to be a real person, and, in my opinion, a native speaker.

TTS is fine for just listening to the odd word or sentence, so if you want to know how to pronounce something for example. I wouldn’t recommend it for this activity.

So look for audiobooks (obviously if there’s an audiobook there’ll be a book somewhere!), or podcasts with transcripts. I’ve recently started using the Whisper function in LingQ to give me a transcript for YouTube videos where there are no subtitles in the original video, and I must say I’m absolutely loving it.

My advice would be: find native speaker audio. Find a written text version somewhere, and if there are none, use AI to generate text. AI text with real native audio is far far far superior to native text with AI audio.

5 Likes