Language Model and Sequence Generation is about audio?

Did I understand correct, that everything that was explained in the video “Language Model and Sequence Generation” on the first week about chances of the wold/sentence was only to resolve the problem of recognition of audio or anything else that doesn’t have visible or correct text input?

What do you mean by ?

Please elaborate:

Andrew names it “chance of the word/sentence”

I meant data that is not clearly written. And can be with errors or like this.

Sorry but I don’t understand your question. Could you please rephrase?

can you clearlify part of my messages that is not clear for you?

I don’t understand anything. Please help me understand your question starting with correcting
the typos in your text.

changed a bit body of the topic. Is not better?

Really sorry but I don’t understand the question.
Other mentors for this specialization have been notified to help you out.

While we wait, could you please fix this part?


Hi @someone555777

it would be great to get some context regards your question. It is quite hard to tell what exactly you trying to ask.

Could you, please, kindly provide some details?


just characterize me, please, the input data, that this video about

So sorry.

I Still doesn’t understand what are you trying to ask.

Andrew NG uses Language model in video. What data does he want to transform with help of this model? What are its characteristics?

So sorry. I Still don’t understand your question.

I’ll assume that you are referring to the examples of the "Language Model and Sequence Generation” slides presentation.

Professor Andrew do not use any data to explain this topic. It is just a hypothetical example of Natural Language Process. It is just text examples.

I hope this help with your question.

Best regards

okay, so we work just with text inputs, thx

1 Like