#voice-recognition

[eletimesdelhi]

New i.MX 8 processors for human machine interface - ELE Times

Arrow Electronics has introduced two out-of-the-box 96Boards platforms to leverage the core performance of NXP’s i.MX 8 processors

[elysiumacademy]

Voice Biometric System – Methods of Hacking and Protecting Measures

Voice Biometric System - Identification of people according to their particular physical or behavioral qualities for example fingerprints. #biometricsystem #voicebiometricsystem #voicerecognition

Little known Methods of Hacking and Protecting Measures @ElysiumAcademy  

[geekychair]

Dragon NaturallySpeaking 14

I just downloaded the newest version and giving it a test run. This is my first post using the software. At this point, I haven't had any issues.

[thethroneroom]

Voice-Recognition Fail

I was all set to sing the praises, in a later post today, of Google’s voice-recognition software (as used in conjunction with its search technology), until the following interaction occurred: What I said - “Marine should always be capitalized” What Google thought I said - “holy shit I was sick at last” Are you kidding me? Where did that even come from? Perhaps the droning of the air-conditioning unit in my office provides more background noise than I thought. FWIW, I moved to a different location and tried it again and the result that time was exactly what I’d hoped for in the first place.

[brainstormtinyvox]

You can't use the robot transcription too in ops with real life conversations without yielding flat out garbage so tape them with tinyvox instead if you want to type them up later..

[bizdojo]

Software Startup "Transcribe Me" Successful in $1.2 Million Funding Round

The winners of last year’s Auckland Startup Weekend, Transcribe Me, have closed an angel investment round, raising $1.2 million. The investment came from a consortium of private investors in New Zealand, the US, and Europe.

Transcribe Me’s SaaS (software as a service) app allows you to record speech, such as lectures and interviews, on your smartphone, and transcribe it into text with a push of a button. A Word document of the recording is sent to you within 48 hours.

What brings an edge to the service is that it combines voice-recognition algorithms with crowd-sourced human transcribers.

A key differentiator for the company is it's use of microtasking to speed up the process. Instead of having one person go through an hour of recording, they split it into 10-second clips and send each clip out to multiple transcribers.

Transcribe Me co-founder Danijel Duvnjak was a member of the team that established our Wellington BizDojo operation last year, and it's fantastic to see him and the rest of the Transcribe Me team achieving such a great result off the back of their Startup Weekend beginnings.

We look forward to hearing more good news as they continue to develop and grow their unique product!

[treatingarthritisinfo-blog]