Comment by davidzweig | Hacker News Reader

Comment by davidzweig | Hacker News Reader

Parent Full thread

I want to hook ChatGPT into Language Reactor (better with realtime ASR and TTS) to make a virtual conversation partner. Latency might be an issue. I was playing a bit with GPT-NeoX before. It would be nice if you could restrict the vocabulary it uses somehow for beginners, if not the range of grammatical concepts. I don't know how you would achieve that. A 'conversation partner' might be a bit aimless as a activity, I thought.. you could feed in the transcript of a short youtube video as a conversation starting point.. or some kind of game..

I just tried this on ChatGPT, looks pretty good:

What is the grammatical error in this sentence: "The dog eat the cake."

>> The grammatical error in this sentence is that the verb "eat" should be in the third person singular form "eats" to agree with the subject "dog." The correct sentence would be: "The dog eats the cake."

What is the grammatical error in this sentance: "I have been living here since three years."

>> The grammatical error in this sentence is that the preposition "since" should be followed by a specific point in time, such as a date or year. The correct sentence would be: "I have been living here for three years." Alternatively, you could say: "I have been living here since 2016."

This also works: Please correct this sentence: "I have been living here since three years." Do not type any additional text.

I’m building something similar for Chinese[0]. Every time a new language model was released, I tested it with pinyin conversion (the standard Mandarin phonetic romanization), and they all failed miserably - until ChatGPT, which so far has performed flawlessly.

Sentence generation isn’t perfect though (at least with my rudimentary prompting) - it will regularly generate “Chinglish” phrases (like word-for-word transliterations from English which are not native Chinese) and occasionally just regurgitate nonsense. It’s not something that I would freely unleash on the end user but there is definitely a place for it somewhere in the pipeline in future.

[0] https://polyvox.app

If you wanted pinyin output, you could use a library like pypinyin? (we use it, no complaints). I assumed GPT models performed best in English, you could machine translate the input/output. MT is pretty fantastic these days, although, you introduce fidelity issues, and an API adds to cost and latency.

I didn't actually need the pinyin conversion (I implement that myself), it was just a test to see how well LLMs* were progressing at "out-of-domain" tasks. I always used recent newspaper headlines, so the possibility of finding matching character/pinyin pairs somewhere in the huge training set was negligible.

For this specific task, Chat-GPT is the first model I've come across that actually shows some ability to chain together concrete steps to form a task (i.e. look up character, fetch pinyin), rather than the simplistic "predict the next word in the sequence" in pure language models.

* Large Language Model - I know ChatGPT isn't a pure language model (which probably explains why it performs so much better), but I don't know what we're calling these models now so I'm sticking with LLM for now.

Figured I'd add a screenshot of my pinyin test with ChatGPT:

https://twitter.com/NickFisherAU/status/1598621215056625664?...

Even if you just had a set of lessons to run through, that'd be a lot of fun. Prompts that make it the counter party in text book dialogues. Character, pinyin, and translation shown.

The email sign up didn't seem to work on your page.

That's basically the idea! Thanks for the note on the email signup, I'll have a look - I threw it together very quickly just before I went on a plane a few weeks ago so it's very fragile.

I've been looking for ages for a way to learn chinese with pinyin (or just audio, no chinese characters), please don't give up.

Thanks! I agree, I'll stay away from characters, at least at an early/beginner level (although I am including them mostly for aesthetic purposes). It's a huge hurdle that lots of people don't necessarily need/want to overcome.

Reply on news.ycombinator.com