Markus Toman

I think it depends a lot on the taco (or whatever) Mel quality. Did you use your own model?

Hi, usually the order of the coefficients is (coeff count - 1)... but actually we have completely different numbers in our scripts... I'll look into this. I've parameterized that huge...

I wonder why they changed that from 47 to 46 in the new version. Just re-exporting would probably be enough to quickly test it, but to be sure I do...

I also don't understand it completely yet, but let's try to :): I just took a look at `tools/make_voice_list`\- so you can use this tool to create a static voice...

Hi all, I just came across this thread because I am watching this repository. Some time ago I (rather) bruteforced German into flite+hts_engine. It was quite painful and messy, so...

I profiled the voice loading once and if I remember correctly the main issue was at https://github.com/MycroftAI/mimic/blob/master/src/cg/cst_cg_map.c#L93 where the mcep trees are read. There are many nested calls reading a...

Ah, I have been at the SSW 2013 (http://ssw8.talp.cat) in Barcelona :). I also trained an HTS version but it turned out to be rather disappointing with the regular hts_engine...

Even if a bit off-topic but perhaps a discussion interesting for others too: Yes, the vocoder is a big bottleneck. I wrote a small tool to do feature extraction and...

> [Full-Band LPCNet: A Real-Time Neural Vocoder for 48 kHz Audio With a CPU](https://xploreqa.ieee.org/document/9455356/) > > GPU is expensive and power-hungry. Pretty bold claim there "The results of these experiments...

@Oktai15 cool, sorry to hijack this a bit but is the NeMo version updated to TalkNet2? Seems the code goes back a year or more where the TalkNet(1) paper was...