A solution for live speech recognition and voice synthesis
Access the utility here (Use Google Chrome): https://huevirtualcreature.github.io/HuesSTTTSSystem/MicrophonePolly
This solution uses Amazon Polly as its voice synthesis service, which I believe should be free up to a few million characters per month. In order to get access to AWS Polly, you need to create an AWS account here. Afterwards, follow these steps (Taken from Here):
-
Sign in to the AWS Management Console and open the Amazon Cognito console at Amazon Web Services Console.-
-
Choose Manage Identity Pools on the console opening page.
-
On the next page, choose Create new identity pool.
-
In the Getting started wizard, type a name for your identity pool in Identity pool name.
-
Choose Enable access to unauthenticated identities.
-
Choose Create Pool.
-
On the next page, choose View Details to see the names of the two IAM roles created for your identity pool. Make a note of the name of the role for unauthenticated identities. You need this name to add the required policy for Amazon Polly.
-
Choose Allow.
-
On the Sample code page, select the Platform of JavaScript. Then, copy or write down the identity pool ID and the Region. You need these values to replace REGION and IDENTITY_POOL_ID in your browser script.
After you create your Amazon Cognito identity pool, you're ready to add permissions for Amazon Polly that are needed by your browser script.
To enable browser script access to Amazon Polly for speech synthesis, use the unauthenticated IAM role created for your Amazon Cognito identity pool. This requires you to add an IAM policy to the role. For more information on IAM roles, see Creating a Role to Delegate Permissions to an AWS Service in the IAM User Guide.
To add an Amazon Polly policy to the IAM role associated with unauthenticated users
-
Sign in to the AWS Management Console and open the IAM console at https://console.aws.amazon.com/iam/
-
In the navigation panel on the left of the page, choose Roles.
-
In the list of IAM roles, click on the link for the unauthenticated identities role previously created by Amazon Cognito.
-
In the Summary page for this role, choose Attach policies.
-
In the Attach Permissions page for this role, find and then select the check box for AmazonPollyFullAccess.
-
Choose Attach policy.
After you create your Amazon Cognito identity pool and add permissions for Amazon Polly to your IAM role for unauthenticated users, you will need two pieces of information for this tool to work: AWS Region key and the AWS Credentials token. You can find this by looking at the Identity Pool ARN.
im sorry that this is so confusing.
When you open the tool, insert your credentials in the upper right text fields and click "Connect".
The keys will be saved to your cookies for this page and will be used every time you visit or reload the page.
Lastly, you need to allow microphone permissions when the page loads. Immediately, the Speech Recognition functionality will start. You may need to reload the page once in order to successfully load the Output Audio Devices (you can check by expanding the dropdown).
Feel free to play with the buttons. You can't really break anything here.
If you want, you can use the sammi extension to turn on/off the tts automatically. You will need to have node installed and be on windows 10 or greater.
Install the sef file and you'll see a new deck. Run the init button once, then restart sammi. Every time sammi connects to obs it will start the server.
When the server is running you can go to http://localhost:61111/ to load the UI (make sure to do this in chrome). You'll need to do the regular set up AND you need to make sure you enable the websocket connection. For the value of the websockets set it to localhost:61112. Then you should be off to the races!