Hello,
As you mention, for Voice applications NXP develops a complete environment that helps customers to control IoT devices using WakeWords.
Here is a fact sheet that will help you to get familiarized with these products.
NXP Software For Voice Processing at the Edge
For your application speech to text NXP does not provide any specific library or product that can help you. However, I think you can use the product VoiceSeeker or VIT API's, these libraries will help you to achieve a great source with acoustic echo cancellation to feed your speech to text conversion library.
In case, you consider using VoiceSeeker with AEC, please contact with the Voice specialized team at the following mail:
voice@nxp.com
Best regards.