Speech-Analytics-API-Examples

Marsview Speech Analytics API Python client end code for Marsview Speech Analytics APIs

Step 1:

Signup on Marsview portal and fetch API Key and API Token Update these values in config.py

Step 2:

A couple of sample configurations for using Speech Analytics APIs are provided in config.py

Shown above is the configuration for running speech-to-text with topics and suggested-intents enabled.

These configurations can be used as is or modified to meet specific requirements. Refer to API docs or more information on how to configure models.

Step 3:

running main.py Note: Modify the Recording file path and the Config. Main.py has 4 stages

• Stage 1: Get Generate JWT Token
• Stage 2: Upload local file and get Transaction_ID (txn_id)
• Stage 3: Send a compute request on the Transaction_ID (txn_id)
• Stage 4: Long poll for the status of the compute request untill the model completes processing.
• Stage 5: Fetch Metadata on the Transaction_ID

Step 4:

Visualise the output using the 'experienceUrl' in the metadata or by visiting Marsview portal