Caffe Deep Learning Framework vs. Nuance Dragon Speech Recognition vs. OpenText Magellan

Caffe is only appropriate for some new beginners who don't want to write any lines of code, just want to use existing models for image recognition, or have some taste of the so-called Deep Learning.

Incentivized

My job requires that I produce lengthy and detailed minutes of meetings and Nuance Dragon Speech Recognition is absolutely ideally suited for this purpose. Notably, meetings are recorded and it is extremely easy to playback the recording of meetings while dictating notes. This is a remarkable saving in time and effort in producing minutes that might otherwise take a few days. I cannot think of any scenario where it would be less appropriate to use Nuance Dragon Speech Recognition other than in a situation where it is not possible to dictate for whatever reason.

Incentivized

If you do not have a large budget and are a large organization, I would steer clear of Actuate. If you are looking to do very complex washboarding, I would not use them. Your developers have to be very skilled to work with this. Plan to bring in consultants if necessary to help your process. Adhoc reporting is weak. If your pricing is user based and you expand, this could be very expensive.

• Caffe is good for traditional image-based CNN as this was its original purpose.

Incentivized

• The main advantage is that it calibrates itself around your tone and accent.
• Integration with many apps and operative system is good.
• Integrating custom words and acronyms is very easy, so, over time, recognition becomes very precise and customized.

Incentivized

• The report outputs can vary across different types such as HTML, PDF, and Excel
• Their open source offering is very sufficient
• There are great boards and blogs for developers and engineers to expand and use their features.
• The people from the company that I've worked with are professional and courteous.

Incentivized

• Caffe's model definition - static configuration files are really painful. Maintaining big configuration files with so many parameters and details of many layers can be a really challenging task.
• Besides imagine and vision (CNN), Caffe also gradually adds some other NN architecture support. It doesn't play well in a recurrent domain, so we have to say variety is a problem.
• Caffe's deployment for production is not easy. The community support and project development all mean it is almost fading out of the market.
• The learning curve is quite steep. Although TensorFlow's is not easy to master either, the reward for Caffe is much less than the TensorFlow can offer.

Incentivized

• Word recognition can always improve.
• Sometimes puts stray words or characters on the page if you forget to tell the microphone to go to sleep.
• Adds non-spoken words for no reason if there is a delay in speaking.

Incentivized

• The documentation on all the available features, but most importantly on the scripting side, can be improved
• The standard look & feel of some basic options, like parameter selection or sorting and filtering, looks dated and can't be customized
• The server portal needs to provide better tools
• More integration is needed with other OpenText products

Incentivized

• Nuance Dragon Speech Recognition is not up to par with any of the other AI voice recognition technologies currently available
• The software is also very buggy and crashes frequently
• Punctuation also seems like an afterthought for Nuance Dragon Speech Recognition software--requiring the user to speak out each comma, period, etc.

Incentivized

I am no longer working for the company that was using Actuate but I believe they would continue to use it because the stitching costs would be to high. It would require a complete rewrite of the reports and the never version of Actuate (BIRT) even required an almost complete report rewrite

Incentivized

Overall, its gives the functionality that I need in my role and can support with automating tasks. I mainly use it for autotext, to add blocks of text and it works universally across all applications. It saves time and works well in Windows 11. It works very well navigating the web.

Incentivized

It is quite intuitive to use. It is fit specifically for doing sentiment, emotion, and intention analysis as well as text classification and text summarization. I would have given 10 if it is fit for the purpose of doing image processing and analysis as well. There is a huge market to analyze video and image data.

Incentivized

One of the worst companies in terms of customer service getting the download file because of a laptop upgrade took me months of work. No jokes.

Incentivized

Always there on the front and backend for us and the client.

TensorFlow is kind of low-level API most suited for those developers who like to control the details, while Keras provides some kind of high-level API for those users who want to boost their project or experiment by reusing most of the existing architecture or models and the accumulated best practice. However, Caffe isn't like either of them so the position for the user is kind of embarrassing.

Incentivized

Other than the more recent speech recognition tools from Microsoft, Google, etc., I have always used Nuance Dragon Speech Recognition. I was introduced to AI technology on an appraisal assignment. During the engagement, I had an opportunity to learn about the technology, and when I researched speech recognition software, the best reviews were of Nuance Dragon Speech Recognition. I purchased Nuance Dragon Speech Recognition and have stayed with the product.

Incentivized

It is vastly superior to these in many ways, for complex reporting it is a much more sophisticated solution. Visualizations are very good. Javascript extensibility is very powerful, others don't support this or as well. Pentaho and MS are both OLAP oriented. Pentaho is moving more toward big data, which was not our primary focus. Others are stuck in the Crystal Reports Band metaphor.

• Since we stopped using Caffe before it can reach the production phase, there is no clear ROI that can be defined.

Incentivized

• Nuance Dragon Speech Recognition has made it easier to dictate progress notes
• Nuance Dragon Speech Recognition has made seeing patients more efficient
• Nuance Dragon Speech Recognition was a little expensive for our smaller practice

Incentivized

• Actuate can handle 50 to 60 sub reports inside a report very well.
• Dynamically creating the datasource, chart, graph, reports are the main advantages. We can do any level of drilling, and can create a performance matrix dashboard efficiently.