Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

Whats the STT Model that DialogflowCX Uses, and how to modify it.

Hi Team currently I have been working with Speech to Text V1 and V2.

When Im setting some specific model on Dialogflow CX for Speech and IVR.

On the model it doesnt specify if its stt v1 or v2

Mizar_0-1743028450426.png

Can anyone let me know where We can get or define that information? I have been reading docs on STT and DFcx and I dont get any insight.

0 4 119
4 REPLIES 4

Hi Mizar,

Welcome to Google Cloud Community!

V1 and V2 is the Speech to Text API version. What you’re seeing is the model use for audio transcription, supported on the Speech-to-Text API used by Dialogflow. See release note for reference.

In case you need to migrate to V2, consider checking this documentation.

Furthermore, I’m not seeing a direct way showing the API version via the Dialogflow CX.

Here’s what you may do instead:

  • Monitor the Speech-To-Text API request. If you have access to API request logs through Google Cloud Audit Logs or Monitoring, look at the actual Speech-to-Text requests being made by Dialogflow CX.
  • You can try testing with v2 parameters.
  • File a feature request through the Google Cloud Platform issue tracker. Specifically request that Dialogflow CX explicitly indicate which Speech-to-Text API version is being used for a given model. While there's no guarantee of immediate changes, this feedback can help improve the user experience and provide the transparency you need. Also, this provides valuable feedback to the product team. You may keep an eye on your submitted feature request to monitor the update.

Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.

Many thanks for your response, so Whats the Version of api for dialogflow? v1 or v2? for STT

Is there anything different there? if this is v1 and v2 then the answer is yes. Or maybe that's all.

Yup there are differences.