Live data from Hacker News

Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

news.ycombinator.com

81–90 of 229 posts

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#83

Earlier quoted context omitted.

I think this is backwards... This is a corpus to train speech to text, not text to speech, right?

It's a corpus designed to capture the full breadth of combinatorial nuances of human speech in a general sense.

No, it is not. For one, it's a corpus of read speech, which means it does not capture well the characteristics of conversational human speech – hesitation, disfluencies, different tones and registers, etc. LibriSpeech has a paper explaining the design of the corpus, all you need to read is the first sentence of the abstract to know what it is supposed to capture:

This paper introduces a new corpus of read English speech, suitable for training and evaluating speech recognition systems.

http://www.danielpovey.com/files/2015_icassp_librispeech.pdf

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#84

IDK about the tech, but I would not worry about it right now. You dont need to play with the tech unless the bad unlikely outcome comes to pass. The only tip I have is from a bit of amateur sound editing I did: collect many samples, and beware of big phrases: Like, ask her to say the same thing many times. And ... sometimes ... to ... stop ... at ... each ... word. And ... so ... me ... ti ... mes at each syllable. O…

> Otherwise, if you ever need to create a sample that contains a single word/syllable, you cant. It is weird how much sound that contains clearly distinguishable syllables for the human ears still is not separable when you go to edit it.

Don’t know why you’re being downvoted. Thought it was insightful.

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#85

I would also suggest looking into learning American Sign Language (of course alongside this project). While communicating via keyboard is workable and good for communicating with the wider world, ASL would be much more convenient for communicating between you two -- and a very interesting language to boot. It is a foreign language thats not related to English besides a few loan words, but there's tons of online resou…

I strongly agree with this. Trying to type never as your main form of communication is exhausting. With a sign language, even if you’re not very good at it, you’re having a face to face conversation and you feel a sense of connection.

Also, if you’re not in America, you can learn your local sign language (e.g. British Sign Language, AusLan)

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#87

Earlier quoted context omitted.

I came here to say this. My brother has a PhD in chemistry and no coding experience. He was able to create a voice model of himself using basic nvidia example generators in a week. My dad lost his voice and it would have been very nice to have a TTS that was much more close to him. I personally would think it would be worth it to have that database. But obviously also attend to the human matters as well, eg spend tim…

I work in pathological speech processing/synthesis so I'm unfortunately familiar with your father's position. It really sucks that these people didn't know that archiving their voice would've been useful. I hear snippets that people manage to glean from family videos right after listening to their current voices and it makes me really sad. On the upside, your father can choose any celebrity he wants to voice him! Ton…

Is Morgan Freeman the most used celebrity?

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#88

Hey, speech ML researcher here. Make sure you have different recordings of different contexts. fifteen.ai's best TTS voices use ~90 min of utterances, some separated by emotion. If you're having her read a text, make sure it's engaging--we do a lot of unconscious voicing when reading aloud. Tbh, if she has a non-Anglophone accent, you're going to need more because the training data is biased towards UK/US speakers. I…

I came here to say this. My brother has a PhD in chemistry and no coding experience. He was able to create a voice model of himself using basic nvidia example generators in a week. My dad lost his voice and it would have been very nice to have a TTS that was much more close to him. I personally would think it would be worth it to have that database. But obviously also attend to the human matters as well, eg spend tim…

Which generator works the best, qualitatively? I come from a vision/ML background but haven't played with speech at all, so it's completely new to me, and wondering what the state of the art is.

I've been wanting to create a TTS of myself so I can take phone calls using headphones and type back what I want to say so that I don't have to yell private information out loud in public locations. Would be nice if during non-COVID times I could sit in a train seat and take phone calls completely silently.

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#89
post #36

Earlier quoted context omitted.

Depends. I'd be horrified to listen to my own voice when I am not speaking. Keep in mind, what we sound very different to ourselves.

> Keep in mind, what we sound very different to ourselves. This is less of a problem with modern high-quality mics than it was, say, with answering machines 30 years ago. Your voice might still sound not exactly the same, but it hopefully shouldn't be unbearably grating either.

Turn the bass up a bit

It's because reproduced audio doesn't have the bass the same as you hearing it conducted through your jawbone (though of course this will sound too bassy to everyone else!)

Re: Ask HN: My wife might lose the ability to speak in 3 weeks – how to prepare?

#90
Here's a simple and practical solution:

Get a decent audio headset, have it record the audio to her phone, and spend hours talking to her about whatever. Preferably in a reasonably quiet environment.

Just spend a lot of time talking. You don't have to talk to her through a headset. Just make sure hers is recording her voice.

It would be easy, painless, and probably good for the relationship too.

Post reply on HN