Speech/Voice recognition

Hey everyone,

I’m trying to find a way to analyze the voice of participants who will be talking in a microphone.

Don’t need it to be super accurate.
I want o be able to trigger different state depending on the the speech.

Women/men
Talking softly
Angry

Such and such

Thanks a lot

Hugues

Moving from Techniques to General.

You can use the Google Cloud Speech-to-Text api if you are just trying to get a transcription of what the user says and run some natural-language analysis on the text:

cloud.google.com/speech-to-text/
cloud.google.com/natural-langua … t-tutorial

Also might be worth looking at Twilio:
twilio.com/speech-recognition

For a recent project we used the windows built-in speech recognition(speech to text, no sentiment analysis or similar). For simple stuff it worked ok and was easy to set up.