Botframework-emulator: How can I use the microphone button at bottom right

[YanXuHappygela]

How can I use the microphone button at emulator and send a voice message? Is there an example of directly sending voice at conversation? I want to use with speech to text and LUIS. Thank you!

[eanders-MS]

I think this blog post describes the process: https://blog.botframework.com/2017/06/26/Speech-To-Text

[YanXuHappygela]

Thank you. But do I have to register my bot to test this out? Or I can simulate this locally at emulator?

[eanders-MS]

No need to register your bot with the bot framework to use the speech services.

__EDIT__: Your bot does need a valid MSA appId/password pair to use speech in the emulator. The simplest way to get one is to register your bot with the Bot Framework. Alternatively, you can generate one yourself at https://apps.dev.microsoft.com

[YanXuHappygela]

Thank you. i have tried to simulate this demo at emulator but it never allows me to input voice or read out the text. Is there any other step I'm missing? To have the communication by voice without registering it as Cortana skill?
https://github.com/Microsoft/BotBuilder-Samples/tree/master/Node/demo-RollerSkill

[EricDahlvang]

Hi @YanXuHappygela

Please try registering your bot on the dev portal, and adding the MicrosoftAppId & MicrosoftAppPassword to the bot's .env ( https://github.com/Microsoft/BotBuilder-Samples/blob/master/Node/demo-RollerSkill/.env ) and to the emulator.

This line: https://github.com/Microsoft/BotFramework-Emulator/blob/c014d0494cd4eaefa5e1b7d1fd6e745c1c2ad504/src/server/conversationManager.ts#L451 in the emulator's code states: 'To use speech, the bot must be registered and using a valid MS AppId and Password.' I'm not sure how to use speech in the emulator otherwise.

I hope this helps. Please let us know how it goes.

[YanXuHappygela]

I have created a MicrosoftAppId and password and created a bot service with this. With this AppId and password, I can use the speech function now! Thank you for your help! I'm curious to know how the app knows that the app will use the speech API? Or it will just use that by default once I have turned on speech to text cognitive service? Thanks a lot!

[eanders-MS]

Thanks @EricDahlvang. Closing this issue now.