In this section, you can configure the Cloud Providers to use in your system:
Here you can create a new Account to use in order to transcribe voice recordings (customers calls that are stored in the Recordings section as MP3 files) and then to analyze the calls transcript (as text files) to detect the sentiment of the conversation.
If your call center often record conversations with customers, it can be very unproductive to have to listen to them every time to analyze how your business is going. Instead, you can automatically use these recordings to improve customer service:
transcribing the speech (using a process called automatic speech recognition) may help you determine the topics that customers are calling about
detecting the sentiment of the conversation (if the call is a compliment or a complaint, that is returning an array of scores to estimate the probability that the input text is positive, negative, neutral, or mixed) can help you implement corrective measures to solve the problem.
Add a New Account
In order to use the functions described above, you need to create a new Account in this section:
Click on the 
button at the bottom right and insert the new Account information
| Panel | ||
|---|---|---|
| ||
|
- Name: a name chosen by you
- Service: the service to be launched to perform the requested automations (see the info box below)
- Specific Service Parameters. In particular for Amazon Services it is requested the following information (you can retrieve them in your Amazon AWS Account):
- Secret Access Key
- Access Key Id
| Info | ||
|---|---|---|
| ||
Up to now, the Account created can use the Amazon AWS services only. Shortly, there will be more services to choose. The Account created use the Amazon AWS services:
This means that in order to make it work, you need to get an Amazon AWS Account. |
Before using the system, please configure in the General Settings the Amazon storage (the nearest Bucket for your geographical region) where data will be saved, the Accounts to be used for both services and the language to be used (only one language per service can be chosen) when you run transcribe and sentiment analysis.