Deep Voice: Real-Time Neural Text-To-Speech
research.baidu.com
Deep Voice: Real-Time Neural Text-To-Speech
1–10 of 80 posts
Re: Deep Voice: Real-Time Neural Text-To-Speech
#2Interesting. They are not TTS like we are accustomed to, they are replicating a specific persons voice with TTS. Listen to the ground-truth recordings at the bottom and then the synthesized versions above. "Fake News" is about to get a lot more compelling when you can make anyone say anything as long as you have some previous recordings of their voice.
Re: Deep Voice: Real-Time Neural Text-To-Speech
#3OK, that went from uncanny valley to flipping amazing. I could picture the person speaking. An old lady. A young woman. It was hard to picture an algorithm in a machine.
It's amazing that is all boils down to 1s and 0s and some boolean logic.
Re: Deep Voice: Real-Time Neural Text-To-Speech
#4does anyone know of good ways to do the opposite, speech to text?
Re: Deep Voice: Real-Time Neural Text-To-Speech
#5does anyone know of good ways to do the opposite, speech to text?
Doesn't google api work for you ? I thought it worked perfect
Re: Deep Voice: Real-Time Neural Text-To-Speech
#6does anyone know of good ways to do the opposite, speech to text?
Depends on how good you're talking. Chrome supports the SpeechRecognition API.
https://developer.mozilla.org/en-US/docs/Web/API/SpeechRecog...
Re: Deep Voice: Real-Time Neural Text-To-Speech
#7how does this stack up against wavenet?
Re: Deep Voice: Real-Time Neural Text-To-Speech
#8Interesting. They are not TTS like we are accustomed to, they are replicating a specific persons voice with TTS. Listen to the ground-truth recordings at the bottom and then the synthesized versions above. "Fake News" is about to get a lot more compelling when you can make anyone say anything as long as you have some previous recordings of their voice.
> you can make anyone say anything as long as you have some previous recordings of their voice.
That's not what this is doing. They're simply resynthesizing exactly what the person said, in the same voice. It's essentially cheating because they can use the real person's inflection. Generating correct inflection is the hardest part of speech synthesis because doing it perfectly requires a complete understanding of the meaning of the text.
The top two are representative of what it sounds like when doing true text to speech. The middle five are just resynthesis of a clip saying the exact same thing. And even in that case, it doesn't always sound good. The fourth one is practically unintelligible. But it's interesting because it demonstrates an upper bound on the quality of the voice synthesis possible with their system given perfect inflection as input.
To clarify, this is cool work, the real-time aspect sounds great, and I'm sure it will lead to even more impressive results in the future. But I don't want people to think that all of the clips on this page represent their current text-to-speech quality.
Re: Deep Voice: Real-Time Neural Text-To-Speech
#9I have A LOT of pdf's I'd much rather listen to than read. Can't wait for this!
Re: Deep Voice: Real-Time Neural Text-To-Speech
#10Interesting. They are not TTS like we are accustomed to, they are replicating a specific persons voice with TTS. Listen to the ground-truth recordings at the bottom and then the synthesized versions above. "Fake News" is about to get a lot more compelling when you can make anyone say anything as long as you have some previous recordings of their voice.
> you can make anyone say anything as long as you have some previous recordings of their voice. That's not what this is doing. They're simply resynthesizing exactly what the person said, in the same voice. It's essentially cheating because they can use the real person's inflection. Generating correct inflection is the hardest part of speech synthesis because doing it perfectly requires a complete understanding of the…
Based upon what little is posted there, I thought they were taking the original recording, then training the model on that recording against the text of the recording... reproducing the recording. I would think next step is to sample enough audio and text to be able to produce new outputs entirely. It should in theory even be able to learn when/where/how to use inflection.