Yoav Ramon

5 Tech trends after COVID-19

2020-04-04T12:00:00+00:00

It might feel far away, and it might take a lot of time, but one day the lockdowns will end, and people will need to go out and continue with their life. The consumers of our world will want to go to restaurants, they’ll like to buy new clothes, and they will need to go to work. In the tech and the startup community, we have to look forward to that day and try to predict what will happen. Here are five trends that I think will characterize this “getting back to normal” period.

Worth noting: These are just my thoughts. Some of them might turn out to be accurate, some of them probably won’t.

You can read the rest of the article at Medium

An updated about section

2019-11-01T13:52:58+00:00

Welcome to my blog!

I wrote more about my interests at the about section, Fill free to contact me!

Lecture on the history and challenges of Diarization

2019-07-02T12:00:00+00:00

This the first time that the “Speech Recognition Israel” community is being gathered and we had more than a 100 participants at that meetup. This is our take at creating a vibrant Israeli community that would be home to share ideas and knowledge about speech recognition, audio processing, and any other speech-related tasks. As a principle, all talks are in Hebrew.

I gave a presentation about the process of diarization. The lecture started with 10 minutes about the historical challenges, continued with 10 minutes about the current solution and finished with 10 minutes about the upcoming challenges. You can find the presentation available online

Speaker Diarization with Kaldi

2019-02-28T12:00:00+00:00

In most real-world scenarios speech does not come in well defined audio segments with only one speaker. In most of the conversations that our algorithms will need to work with, people will interrupt each other and cutting the audio between sentences won’t be a trivial task.

In addition to that, in many applications we will want to identify multiple speakers in a conversation, for example when writing a protocol of a meeting. For such occasions, identifying the different speakers and connect different sentences under the same speaker is a critical task.

Speaker Diarization is the solution for those problems. With this process we can divide an input audio into segments according to the speaker’s identity. It can be described as the question “who spoke when?” in an audio segment.

You can read the rest of the article at Medium

How to start with Kaldi and Speech Recognition

2018-10-22T12:00:00+00:00

Kaldi is an open source toolkit made for dealing with speech data. it’s being used in voice-related applications mostly for speech recognition but also for other tasks — like speaker recognition and speaker diarisation. The toolkit is already pretty old (around 7 years old) but is still constantly updated and further developed by a pretty large community. Kaldi is widely adopted both in Academia (400+ citations in 2015) and industry.

Kaldi is written mainly in C/C++, but the toolkit is wrapped with Bash and Python scripts. For basic usage this wrapping spares the need to get in too deep in the source code. Over the course of the last 5 months I learned about the toolkit and about using it. The goal of this article is to guide you through that process and give you the materials that helped me the most. See it as a shortcut.

You can read the rest of the article at Medium