Comments (10)
Well, the Spanish support will hopefully land really soon (hopefully this week). And French would be next, but probably not that soon...
It seems we have found a high quality dataset of recordings for creating a speech model in French, but we are still lacking a good language model.
In particular we are looking for a free software tool (preferably MIT or BSD-like licensed) able to phonetically transcribe and syllabify French sentences. There are some GPL options we may end up using instead.
from mimic1.
The initial version still has a lot of room for improvement, but I got some sort of audio at least from Zoé, our first French voice:
There is a lot of work to do in many aspects. Currently the voice speaks too fast and pauses are not well inserted, but as something I've done on my free time is not that bad!
There is a lot of work in the word->phoneme dictionary in order to get good pronunciation. I created the entries with a quick and dirty script and I did not review it manually. I know there are duplicated entries, some entries with wrong transcriptions, extra or missing phonemes (especially at word boundaries), etc...
We are dividing mimic into several modules and repositories. Until some pull requests are reviewed and some packaging work is done it won't be straightforward to try the French voice.
Once we have an easy way to try the French voice, if someone with French skills and time wants to start adding and fixing entries, they will simply need to fork the https://github.com/MycroftAI/mimic-french repository, edit the file https://github.com/MycroftAI/mimic-french/blob/development/src/lex/fr/fr_dict.out and submit pull requests. Of course you can already do this, but you won't have an easy way to see the improvements of your contributions right now.
from mimic1.
Do you need help with french-speaker voice files? Or is there an algomagic that can adapt any voice to any language?
from mimic1.
Any help is appreciated I'd say. @LongBoolean may correct me if I'm wrong but as far as I can tell voices and languages are somewhat separated in this kind of system. To add french a voice I think the quickest path would be using an existing voice and adding new language pronunciation rules.
If you'd like to work on this there are some documentation found on various places around the net:
http://www.festvox.org/flite/doc/flite_8.html
https://sourceforge.net/p/at-flite/wiki/AddingNewLanguage/
from mimic1.
A dataset of speech recordings in French aimed to build TTS systems.
http://datashare.is.ed.ac.uk/handle/10283/2353
from mimic1.
Nobody knows when will add the French ?
from mimic1.
I am pleased to read that the integration of French is still current
I would wait ! Hoping that the integration of French is fluid and clear, even with the management of accents and intonation.
Good luck
from mimic1.
Really impressive! Without disconcerting, everything begins somewhere not true?
For now the voice is not really human, the timbre is robotic, the connections ("liaisons" in french) are not respected, in short, there is still work to do ! I am super excited about what we could build with a French voice! So much opportunity !
If I have the time, I would like to build a NodeJS interface => Mimic (with the Matrix Voice project, it would be just incredible!) 👍
Thank you for your work and your project
from mimic1.
Bonjour, le projet d'une voix mimic Française de bonne qualité est toujours d'actualité ? Existe t-il déjà quelque chose d’intéressant ? Merci !
from mimic1.
Hello, the project of a voice mimic French of good quality is still relevant? Is there anything interesting already? Thank you !
from mimic1.
Related Issues (20)
- Publish prebuilt mimic releases HOT 2
- windows cross-compilation instructions are not exact. HOT 2
- Problem with speaking. HOT 7
- Voice is not playing over bluetooth HOT 3
- Where is the difference between Mimic and Flite? HOT 2
- Linking fails, but easily fixed HOT 2
- Add support for Palm OS HOT 2
- Unable to make standalone mimic pause and resume at will when reading long text from terminal HOT 5
- Compile for pulseaudio HOT 2
- Using mimic libraries HOT 11
- Distorted 'ap' voice output when compiling with gcc 8.3
- mimic -p '' segs HOT 1
- Tool for creating my own .flitevox? HOT 4
- Dead links: festvox.org doesn't work anymore? HOT 3
- Is there a guide for adding new language support to mimic1? HOT 1
- Windows cross compilation error
- Issue while compiling with LTO enabled HOT 1
- Error when running 'make' HOT 5
- At ./autogen.sh: syntax error near unexpected token `newline'
- "Shared" compilation (with --enable-shared=yes) fails. HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mimic1.