asked on

Need Speech to Text service

Need Speech to Text service

I have an iPhone 6, a MacBook and a Windows PC. I need to find a service which lets me speak into some device and get the text version emailed to me. I do not have time for a service, which can cost $1/minute and take 12-24 hours. I need it immediately, which means the text creation needs to be near real time.

I will pay for this service, if needed. I think $1/minute is the highest I could pay.

by "some device" I meant:
1) My iPhone 6
2) A microphone hooked into a Mac
2) A microphone hooked into Windows

Thanks,
Robert

ASKER CERTIFIED SOLUTION

David Favor

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

SOLUTION

Michael Painter

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

curiouswebster

ASKER

I have heard Dragon is nowhere as good as AWS Transcribe API. But, once you drop the $150 or Dragon, you're stuck. I have also read that Google Cloud Speech API is very good.

I need the best dictation possible since this is for a guy who has no use of one of his right hand. So, I am searching for options before I ever think about creating a speech to text front-end website.

What service does your Android use? Is the translation as good as Google Cloud Speech API? This service costs about $1.5/hour of spoken audio. And idea how often there is a mistake?

I have an Android Note4, so maybe I could try that out...It uses a very old OS...

Do you need Keep for the translation? How do I access from an Android?

curiouswebster

ASKER

thanks

David Favor

For completeness...

https://github.com/benoitfragit/google2ubuntu - Command line hook for Google speech to text API V2.

https://www.ubuntupit.com/best-open-source-speech-recognition-tools-for-linux/ - With special mention of Julius, which is a bit complex to get working. Julius is suitable when 1000s+ files require continuous conversion frequently, say for hourly recordings of ambient sound or for podcasts or video channels producing massive content.