Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
Can someone give me an intuitive explanation for why we use word embeddings and how neural networks make this process easier and better?
Word embedding is the collective name for a set of language modeling and feature learning techniques in natural language processing (NLP) where words or phrases from the vocabulary are mapped to vectors of real numbers.
There are many branches and many research groups working on word embeddings. In 2013, a team at Google led by Tomas Mikolov created word2vec, a word embedding toolkit which can train vector space models faster than the previous approaches.Most new word embedding techniques rely on a neural network architecture instead of more traditional n-gram models and unsupervised learning.
Why do newer methods rely on neural networks? Mostly because they can solve the problem faster and more efficiently. Part of the reason for this is neural architectures take the whole sorted sequence into account, and not each word in isolation.
For a more detailed explanation, you probably want to do some research. Reading this blog post might be a good start.
Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 1 year ago.
Improve this question
this is more of an hypothetical question, but might have great consequences. Lot of us in Modelica community are dealing with large scale systems with expensive simulation times. This is usually not an obstacle for bugfixing and development, but speeding up the simulation might allow for better and faster optimizations.
Recently I came across Modia possibilities, claiming to have superb numerical solvers, achieving better simulation times than Dymola, a state-of-the-art Modelica compiler. The syntax seemed to cover all important bits. Recreating large scale component models in Modia is unfeasible, but what about automatically translating the flattenized Modelica to Modia? Is that realistic? Would that provide a speed up? Has anyone tried before? I have searched for some
This might also hopefully improve integration of Modelica models and postprocesssing / identificaiton tooling within one language, instead of using FMI or invoking a separate executable.
Thanks for any suggestions.
For those interested, we might as well start developing this.
We in the Modia team agrees that the modeling know how in Modelica libraries must be reused. So we are working on a translator (brief details given in https://ep.liu.se/ecp/157/060/ecp19157060.pdf) from Modelica to Modia. The plan is to initially provide translated versions of Modelica.Blocks, Modelica.Electrical.Analog and Modelica.Mechanics together with Modia.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 6 years ago.
Improve this question
Recently, I was reading a paper whose name is "On the Effectiveness of Concern Metrics to Detect Code Smells: An Empirical Study".
I come from a non-English speaking country, and I can not quite understand what Concern Metrics means in the field of software engineering.
It is not referring to the relationship between objects?
I have some understanding of java and c #, some people may be able to use java to give me an example.
Thanks.
Like it is said in the paper's abstract: "While traditional metrics quantify properties of software modules, concern metrics quantify concern properties, such as scattering and tangling." Are you familiar to the cross-cutting concern concept? This question provides examples of concerns: Cross cutting concern example Try to read papers on aspect-oriented programming (AOP) to grasp more concepts in order to understand better the relationship between concerns and code. The metrics are attempts to quantify, for instance, the amount of scatterness of a concern (e.g. login) over the source code.
Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 6 years ago.
Improve this question
Recently, I am working on my course project, the topic is the creation of a new hybrid software process model by integrating Scrum and Team Software Process (TSP). Integration of these two models will be based on the SEMAT Essence Kernel Framework.
I am wondering:
Which steps should be followed for this integration (like
determination of the roles and artifacts in these two models)?
What should be the criteria to decide on good sides?
Thanks in advance!
The best way I think I can answer this question is by quoting the agile manifesto.
"Individuals and interactions over processes and tools"
Agile is about people, teamwork and craftsmanship. It's about involving the customer closely to figure out what really is needed - and delivering that, in small increments of working software. Agile is inspect and adapt, based on experimental delivery and the feedback and evidence that comes from that.
Trust yourself. Work closely together and you can do this. The best learning often comes from doing. :)
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
What can you do after learning big data related concepts, Hadoop, ML, NLP etc? Where can you implement these?
Not really a software related question - but it's very relevant to current technology and why some software exists. So here is an opinion.
We now live in a world where it is possible to monitor and digitally record information on an epic scale that continues to expand with concepts like The Internet of Things.
With this information it becomes possible to look at the evidence behind decisions that previously would have been made by gut instinct or opinions. What impact does road design have on traffic flow? Which medical drugs get best results in the real world and not just in drugs trials? Is there a correlation between office temperature and productivity? and so on for millions of questions in different domains.
Around the world, organisations are using data they never previously had, to get better at whatever they do (Good or bad).
The big data concepts are the tools for managing all this information. Big Data is not just large in volume, it is often unstructured and in different forms.
So to answer your question. You can implement these concepts by working with organisations that are using Big Data. Hopefully you can see the potential of Big Data along with the mind bending headache it can create when trying to make sense of it.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 1 year ago.
Improve this question
Is it possible, programatically, to take someone's voice sample and produce a unique tone/property that could be used to create a synthesised speech?
For example, person A records himself. A unique tone is produced from this voice sample, and is being turned into synthesis speech. This allows people to use this synthetic voice in Text-to-Speech software, writing any text that they want that would be read in person A's voice.
Is it possible in today's terms? I know that there are companies that do this professionally, but generally, is it possible for a piece of software to do this?
Using speaker adaptation methods you can achieve some results with comparably few training samples but still you should have some hundred sentences of the person - preferably with a phonetic transcription.
We once had this as a small lab exercise for students to record their own voices and train a voice model using HTS (http://hts.sp.nitech.ac.jp/).
The "most simple" approach using HTS is to download the "Speaker dependent training demo" from this page and replace the training speech samples with your own recordings (of the same sentences!).
We did this for another language with our own package though.
I think MaryTTS (http://mary.dfki.de/) has some more convenient tools to assist with this process but I've never worked with that.
But still - for high quality voices, you should have thousands of recorded sentences.
In 2021 and beyond I suggest to use mozilla/tts which is the best if you want to step in and use an existing, proven stack.
Seven years later, you can use your voice for text-to-speech:
Overdub: Ultra realistic text to speech voice cloning
https://www.descript.com/overdub
There was a Bloomberg documentary about "Lyrebird", a neural network that can learn your voice and then you can make new sentences with it. Descript was founded by the people who made Lyrebird and now offers this service, and non-linear editing for synthesized audio.
Link to the Bloomberg documentary on YouTube: https://www.youtube.com/watch?v=VnFC-s2nOtI