The Deepspeech server is serving STT/TTS. I just don’t think it will run another model type.
Deepspeech just does STT. Tacotron is TTS.
This is about the later.
@baconator
Oh OK, now i’ve dug a little deeper i saw that the articles talking about 2 servers with the second model already packaged, so i haven’t recognized it as such.
So, STT aside. Is the TTS serveing (described in the how-to) still viable? Or what would you suggest?
Is STT modeling sourced from one speaker beneficial?
If you’re referring to TTS/TTS/server at master · mozilla/TTS · GitHub then yes, this is still viable and I just sent off a package earlier today using that.
Possibly for that one person. A wider set of submitted data would almost certainly help, even if the bulk of the data was from one person.
We first want to iron out the shortcomings mentioned above, e.g. “stop attention” and voice quality. After that the TTS & vocoder models will be published.
2 Likes
As discussed in Mycroft chat with @SGee i’ve uploaded the sample phrases as in first post with a new “vocoder” (wavegrad) model training. @Dominik and i are currently playing around with different vocoders.
It’s based on same taco2 model as first samples (460k steps), so voice flow is identical but it’s pronounced diffently. Random noise in background will (hopefully) get away on more training steps (currently wavegrad training on 350k steps).
3 Likes
My german offline TTS model is ready to be used in Mycroft.
Here’s a video showing how to set it up.
4 Likes
thx a lot for all the hard work. works great but sadly on my machines quite slow and therefore almost not usable since I’m not sure how to improve the time it takes to generate the wav.
A better machine. What are you trying it on now?
Intel® Xeon® CPU E3-1246 v3 @ 3.50GHz
or a
CPU(s) 8 x Intel(R) Core™ i7-6700T CPU @ 2.80GHz
Got a GPU on either one?
no… just the integrated ones in the CPU ( Intel® Xeon® CPU E3-1246 v3 @ 3.50GHz)
I’d check if the model can be run easily with the Griffin-Lim vocoder. That could be faster but with less quality.
I’ll check and give feedback if i know it’s working.
thx a lot. In our video what kind of hardware did you use, the developer kit is based on what chip?
Guude,
@Dominik and i are still working on the next release of “Thorsten” voice to be used as offline TTS for Mycroft. (@Olaf thanks for supporting with HifiGAN training compute power)
Some german sentences taken from Mycroft skills can be heard here:
3 Likes
Even if this thread a a little outdated because of @synesthesiam fantasic work on Mimic 3, but i still work on providing a free, high(er) quality TTS voice.
I’ve trained a new model and am unsure which of these two variations to be released.
Please give the samples a listen and let me know which variation you like more.
WOW it sounds really great!
Is there a possibility to run it as a docker?
So we can acces it over http://localhost:5002/api/tts?text=Hallo%20wie%20geht%20es%20dir
How about providing a Plugin for Openvoice OS TTS engines with whatever your model supports as a Text to Speech engine?
I have the newest (.8.0) coqui-tts docker successfully deployed (gpu-enabled). Can’t speak for cpu only, yet this should be straight forward with the given Dockerfile.
GPU on the other hand needs, if nvcr.io/nvidia/pytorch:22.08-py3 is used, changes to the Dockerfile and setup.py directly in the source code. I only got it to work with a conda install (nvidia uses conda), otherwise pytorch will throw exceptions left and right.
The plugin for the Coqui server is in the making. (But you would need to build coqui tts form a fork of mine, it needs an api addition and the possibility to define a conf.json loaded at startup)
1 Like
thx a lot.
At least I got the docker running but I’m not sure how to build the docker to use TTS_VERSION=0.8.0 like he mentions here: