Speech model migration for phone_call in Q2 2024

Starting April 17, 2024, for certain language tag and speech setting combinations, the Dialogflow CX and Dialogflow ES traffic with audio will gradually route away from the classic Speech-to-Text model behind the phone_call model identifier to the new conformer-based speech models. As the redirection will be internal only, the current model identifiers and code will remain valid and continue to serve traffic.

The affected language tags are: en-au, en-gb, de, de-de, es, es-es, es-us, fr, fr-ca, fr-fr, it, it-it, ja, pt-br.

The following table shows the routing of Speech-to-Text phone_call model that will take place when the migration takes effect. If your configuration matches the language tag and the first three columns in the table, the new model identifier is listed in the fourth column.

Current model identifier Current model_variant Current single_utterance setting * Model identifier that traffic will be directed to
phone_call USE_STANDARD false latest_long
phone_call Other than USE_STANDARD false telephony
phone_call any true latest_short

Migration options

There are three migration options: early opt-in, automatic, and temporary opt-out.

Opt in and migrate earlier

To opt in proactively, change the speech model identifier that you have been using in the Dialogflow CX API and Dialogflow ES API with the updated one. For Dialogflow CX, you can also change the model identifier in the Advanced speech settings under Speech and IVR settings in the Dialogflow CX console.

Speech model specified in the API at runtime takes precedence over the speech model specified in the Dialogflow CX console, unless you enable the Override request-level speech model setting. Read more about precedence in speech models.

For partner telephony integrations, check with your telephony provider if the speech model is being set in the API. If you are using the Dialogflow CX Phone Gateway, review the speech model limitations.

Migrating your project proactively gives you time to test the models and take advantage of the improved accuracy and robustness earlier. Most integrations require single_utterance = true, so if you are not using advanced timeout-based end of speech sensitivity you'll likely want to use latest_short.

Migrate automatically

To be migrated automatically, no action is required on your part. The existing Speech-to-Text model identifiers are valid and we will shift traffic internally.

Opt out and migrate later

If you find any issues with the updated models and would like to opt out from the migration temporarily, raise a Google Cloud support case by April 17, 2024. When creating the support case, use the title "Opt out from Dialogflow CX Speech-to-Text migration" and provide your project IDs and the reason for opting out. Opt-out requests won't be accepted after April 17, 2024.