PieSquared

6 ms·

1.
Facebook's Knowledge-Assisted Neural NLP Papers of 2020(andrew.gibiansky.com)
2 points by PieSquared 6y ago|0 comments
2.
Understanding DiffWave and WaveGrad, New Neural Speech Synthesizers(andrew.gibiansky.com)
3 points by PieSquared 6y ago|0 comments
3.
by PieSquared 6y ago
Probably worth adjusting the title -- air quality is not just a measure of pollution. In this case, both SF and Portland are filled (unfortunately) with wildfire smoke, rather than a man-made pollutant...
4.
by PieSquared 7y ago
Are you folks planning on extending this to speech? I'm always been disappointed by how speech vocoder networks aren't built with any great inductive biases for waveform generation (besides very long receptive fields), and have de
5.
by PieSquared 7y ago
I don't think anything about the current set of tools is specific to sample rate; WaveNet, Tacotron, WaveRNN, etc, should work fine to generate 44.1Khz audio. They might just need slightly different hyperparameters or sizes to work wel
6.
by PieSquared 7y ago
There's a few recent papers actually that show minor improvements by integrating LPC prediction into deep methods ([0], [1]). In my experience (some of which comes from reproducing these, some of which comes from my own experiments), t
7.
by PieSquared 7y ago
First of all, it's important to note that Tacotron and WaveNet are responsible for different parts of the speech synthesis pipeline, so the comparison here isn't quite accurate. Specifically, Tacotron takes a representation of the
8.
by PieSquared 7y ago
I'll try to answer these one at a time. 1. Does text-to-speech require AI? This one is a bit tricky to answer since it requires defining "AI". AI as a moniker has been used to describe deep neural networks, search algorithms,
9.
by PieSquared 7y ago
I'm an author on a few of these papers referenced (the Deep Voice papers from Baidu). I'm happy to answer any questions folks may have about neural speech synthesis, as I've been working on this for several years now. In gene
10.
by PieSquared 8y ago
Ta-Nehisi Coates, Between the World and Me. https://en.wikipedia.org/wiki/Between_the_World_and_Me "It is written as a letter to the author's teenage son about the feelings, symbolism, and realities associate
11.
by PieSquared 8y ago
The new Gmail interface has snooze, so if that's the only thing you're missing it'll stick around. I mostly miss the "Trips" feature, having recently switched from Inbox to new Gmail...
12.
by PieSquared 9y ago
"Monoid" is an adjective that describes a data type. Anything you describe as a monoid has to have three properties: you can add them together, there's an "empty" or "zero" value, and (a + b) + c = a + (b
13.
by PieSquared 9y ago
In addition to the blog post, there's an interesting discussion on Github that happened before the blog post was published: https://github.com/Unikernel-Systems/unikernel.org/pull/45
14.
by PieSquared 10y ago
To be fair, this article is really pretty "cutting edge" as far as Haskell goes. It describing a new feature they're planning on adding to the Haskell compiler, one that significantly extends the type system in an interesting
15.
Baidu launches SwiftScribe, an app that transcribes audio with AI(venturebeat.com)
2 points by PieSquared 10y ago|0 comments
16.
by PieSquared 10y ago
Feel free to get in touch for more Q/A, my email is in my profile. We've experimented a bunch with many of these hyperparameters. Our phoneme signal has mostly stayed 256 Hz, but we've done a few experiments with lower-freque
17.
by PieSquared 10y ago
I don't quite know what VoCo does, but it seems like a concatenative system that they've tuned a huge amount. I'm a little skeptical that it works as well and as reliably in real life as it does in demos. But, even so, there
18.
by PieSquared 10y ago
Check out Char2Wav (recent) and SampleRNN (the RNN-based audio synthesis architecture). The related work section of the Deep Voice paper mention a bunch of related papers that are relevant!
19.
by PieSquared 10y ago
We take several days (2-3) on 8 Titan X GPUs to train our models, which is quite a lot of compute. Running on mobile devices is quite challenging – the inference is not yet fast enough to support that, and has only been optimized for x86 AV
20.
by PieSquared 10y ago
It's hard to say! We don't quite know exactly how many parameters or minutes of audio are needed to describe fully someone's voice and speaking patterns. Maybe one or two, maybe much more.
21.
by PieSquared 10y ago
Thank you for clarifying this! We tried fairly hard to make this clear, because as you say, the hard part is generating inflection and duration that sounds natural. There's still a ton of work left to do in this duration – we're c
22.
by PieSquared 10y ago
Right now, we do not have plans to make an API available. This paper and blog post are mostly meant to describe our techniques to other deep learning researchers and spur innovation in the field. However, we hope that these techniques will
23.
by PieSquared 10y ago
Baidu as a company doesn't use TensorFlow (as far as I know). We have our own high-performance and easy to use open source framework called PaddlePaddle [0], which is quite powerful and flexible. However, the Baidu Research Silicon Val
24.
by PieSquared 10y ago
Thank you! For fully end-to-end models, it's hard to say exactly. The Char2Wav paper demonstrates that there is hypothetically an architecture and a set of weights that can do synthesis end-to-end, but we cannot yet train such a system
25.
by PieSquared 10y ago
We didn't actually try LSTMs, because we train in 1.25 second chunks, so running an LSTM for several hundred timesteps would drastically slow down training. Our per iteration time was in the 200-500 milliseconds, and using an LSTM or G
26.
by PieSquared 10y ago
We are not currently releasing any code, but hopefully the paper on arxiv is enough to make it easy to reproduce the result. We use TensorFlow for writing and training the model and c++ with a lot of hand optimizations for inference, with a
27.
by PieSquared 10y ago
Hey there! I'm one of the authors of the paper and I'm happy to answer any questions anyone may have! Make sure to check out the paper on arxiv as well.
28.
by PieSquared 10y ago
Hey there! Our paper is on Arxiv [0] and we have a blog post about this [1]. I am one of the authors on this paper and am happy to answer any questions you may have. [0] https://arxiv.org/abs/1702.07825 [1] http:&#x2F
29.
by PieSquared 10y ago
For some reason, either my account or the the baidu.com domain is blocked or delayed on HN. We tried to post this but weren't able to post the link.
30.
Deep Voice: Real-Time Neural Text-To-Speech(research.baidu.com)
244 points by PieSquared 10y ago|77 comments

More ›