sleepwalking / shiro Goto Github PK
View Code? Open in Web Editor NEWPhoneme-to-speech alignment toolkit based on liblrhsmm
License: GNU General Public License v3.0
Phoneme-to-speech alignment toolkit based on liblrhsmm
License: GNU General Public License v3.0
hello, first thanks for the nice framework.
the extraction of mfcc and first, second-order delta feature works well.
After that, when I load the model(.hsmm)
the Error: failed to load model from blah blah
.. error is occurred.
Some model file(empty.hsmm) doesn't occur above error.
And i made some test.txt or text.hsmm file and change the from path to test file to check the fopen function in hsmm = load_model(optarg) in shiro-rest.c whether it works well. But it also got an error!
fopen return success by checking 'perror', it returns 'Success'. the custom c file i made also can read any .hsmm and test.txt.
but it doesn't works only in your shiro-rest.c code.
I can't resolve this situation, how can i resolve this problem?
Hello,
I am building a dataset to train with and need to ask a few questions before proceeding.
What is the max supported/suggested audio length? is several minutes alright or should the audio be limited to about ~20 seconds or so? Likewise, is there a reasonable limit to the length of the index?
Thank you.
Hi,
Building on linux, I'm encountering a problem running SHIRO.
I've tried with adding the .lua to the extrator as well but I get the same error.
lua5.2 shiro-fextr.lua ~/Downloads/UTAU/Resonance_Harmony_Arpasing_English/Base_B3/index.csv -d ~/Downloads/UTAU/Resonance_Harmony_Arpasing_English/Base_B3/ -x ~/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k -r 16000
lua5.2: shiro-fextr.lua:54: module '/home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k' not found:
no field package.preload['/home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k']
no file '/usr/local/share/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.lua'
no file '/usr/local/share/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k/init.lua'
no file '/usr/local/lib/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.lua'
no file '/usr/local/lib/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k/init.lua'
no file '/usr/share/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.lua'
no file '/usr/share/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k/init.lua'
no file './/home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.lua'
no file '/usr/local/lib/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.so'
no file '/usr/lib/x86_64-linux-gnu/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.so'
no file '/usr/lib/lua/5.2//home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.so'
no file '/usr/local/lib/lua/5.2/loadall.so'
no file './/home/myname/Prog/audio/SHIRO/extractors/extractor-xxcc-mfcc12-da-16k.so'
stack traceback:
[C]: in function 'require'
shiro-fextr.lua:54: in main chunk
[C]: in ?
I am looking to use Shiro to label speech with the stresses in-place. Does Shiro have support for this without treating them as a unique phoneme?
If not then would it be ok to request this as a feature? Being able to do something like ah durfloor 0.4 aka ah0 aka ah1
as to not waste data but still output the stress in the final label would be very useful.
Thank you.
Hi,
lua shiro-fextr.lua index.csv -d "../cmu_us_bdl_arctic/orig/" -x ./extractors/extractor-xxcc-mfcc12-da-16k -r 16000
Can you tell what is the content of index.csv file which is one of the input argument for speech-phoneme alignment.
Also what path should be provided for -d argument
Thanks
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.