question

samy-adv avatar image
samy-adv asked ·

Use Alexa as Speech-to-text engine

Can we use Alexa as speech to text engine? Is there any workaround though Skill and get the text of my spoken words?
alexa voice service
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Gautam@Amazon avatar image
Gautam@Amazon answered ·
Hello Samy, Alexa cannot be used as a generic speech-to-text engine. If you create a Skill, then you will be able to get the text of spoken words, but only for your Skill's slots. I hope this answers your questions. ~ Gautam
1 comment
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Hey Gautam, I had a query regarding why does alexa sills does not allow speech-to-text and get the spoken words.

0 Likes 0 ·
samy-adv avatar image
samy-adv answered ·
"get the text of spoken words" ... can you connect me to document or example of that? Also is there any list of Alexa REST APIs which developers can browse?
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Gautam@Amazon avatar image
Gautam@Amazon answered ·
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Birkan U. avatar image
Birkan U. answered ·
What is the recommended way of sending the spoken text from the intent to our own service? Websockets or HTTP? If so, how do you make a connection?
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Levon@Amazon avatar image
Levon@Amazon answered ·
Hi Birkan, Thanks for posting. Please refer to the links in Gautam's answer above: "Getting Started" guide for Alexa Skills: https://developer.amazon.com/public/solutions/alexa/alexa-skills-kit/getting-started-guide Specifically in the 'Getting Started' guide, go through the 'Hosting as a Web Service on Your Own Endpoint' section, as it contains all the details you need for your implementation. Kind regards, Levon
10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

N3wbi3 avatar image
N3wbi3 answered ·

Do I understand this right? There is no way, that i send the hole spoken sentence via an intent to my service? The only way is to implement a slot in an intent. And only the words that are defined this slot are send via a json object to my servie. I only ask, because i would like the parse the spoken sentence by my selfe. In this way my service would be more easier to implement and more flexible.

Sorry that i start this topic again, but maybe it changed over the time. Thank you guys.

10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

carbrain-dev avatar image
carbrain-dev answered ·

Hello I'm interested in the speech-to-text with Alexa, so can you confirm the only way to grab the test is to send it to my local server from a custom skill? Is there any Amazon API available that allow me to capture my queries like those shown in the Settings->General->History section at alexa site?

10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Sanjay@Amazon avatar image
Sanjay@Amazon answered ·

Btw, another way you can to speech-to-text is using AWS Lex (https://aws.amazon.com/lex/).

If you want to stick with Alexa, I'd recommend posting this question in the Alexa Skills Kit forum to get better advice on how to use a skill to do this specifically.

10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Sanjay@Amazon avatar image
Sanjay@Amazon answered ·

Btw, another way you can to speech-to-text is using AWS Lex (https://aws.amazon.com/lex/).

If you want to stick with Alexa, I'd recommend posting this question in the Alexa Skills Kit forum to get better advice on how to use a skill to do this specifically.

10 |2000 characters needed characters left characters exceeded

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.