Do you spend too much time browsing audio records looking for a key insight? Have you ever wished you could make a deep analysis of the best wording to close a sale? Ever wonder how programmable communication & speech to text could help increase the value of your voice channels?
This article will discuss how Wazo Enterprise Unified Communication could increase the value of voice channels. More precisely, we will investigate the use of speech to text service. But you could use any service that can transform, analyze, and enrich your voice channels.
Deploy Speech To Text to enrich your voice channels
ASR (Automatic Speech Recognition), STT (Speech To Text) or speak engines all refer to the same technology. The aim is to be able to transform voice, stream or file, to a text. Text remains the best medium to leverage modern technologies such as AI, indexing, and searching.
Imagine you could provide accessibility features within your application. For example, a phone call being transcript in real time. A partially deaf person could read what the speaker is saying and answer it by typing. The other person being read the answers by a vocal synthesis (Text To Speech) service. You could bring an inclusive and smooth communication experience to all users.
Imagine having not only audio files, but full call center conversation transcripts injected in the related Salesforce chatter thread. You would be able to search and analyze with precision whatever you want. Making advanced linguistic-based statistics or easing legal matter process. Creating a brand new definition of operational efficiency.
Another example. You are receiving too many vocal messages and don’t have time to deal with it. You could receive it transcribed in a Slack or Microsoft Teams channel, an e-Mail or even SMS. Enabling your people to gauge its value and take quick actions. Increasing your team collaboration and productivity.
Wazo Enterprise Unified Communication provides all the APIs to build such use cases and many other telecom scenarios. Using a third-party solution like the free visual builder Node-Red, you could build your own flow answering your business needs in a couple of clicks.
Using programmable voice to transcript voice messages in a business message system
Here is the workflow that describes the different steps to enrich your voice channel with Speech to Text:
First, we will receive a call. This call will be routed to a voicemail.
Next, our audio data will have to be injected in our Speech-To-Text solution which will return us transcribed data.
Finally, these data will be enriched and formatted to be sent through various media, including e-Mail. Mission accomplished!
Let’s go deeper in the tech
In this example, we used Wazo Enterprise Unified Communication, Slack, MatterMost an SMS gateway and Gmail.
First, let us give you some insight about our fetch voicemail node. We are using this node to retrieve audio data from the user’s voicemail. But it also includes a WebSocket linked to the new audio message event. In other words, it will act as a trigger for this scenario every time a new voice message appears in the user’s voicemail. It will provide us audio data in 2 different formats. On the one hand, an audio message file and on the other hand the same message in a binary data format. This last format we will be stored in the Node-RED built-in buffer thanks to our “Get user” function.
Now we are injecting these binary data in our Speech to Text service provider which will return us a text format.
And finally, all we need is to send this data through our different channels. For e-Mail and SMS channels, we will ask our Wazo Engine to give us this information using a GET method on our “Confd” API. Then we use third-party nodes to connect to each channel and send our transcribed data
Are you looking for help understanding Wazo’s API possibilities? Please find our related documentation: here.
Discover additional programmable communications capabilities
Using Node-Red, you are able to solve complex business challenges, combining Wazo and other software, without compromising with users’ needs.
If you want to learn more about programmable communications, we built additional use cases for you:
- “Easily design a voice alerting system” – Episode #1
- “Build an audio conferencing solution” – Episode #2
- “Turn your communication data into business analytics” – Episode #4
- “An IVR system built for your needs” – Episode #5
If you feel that programmable communication can help you solve your business challenges, feel free to reach out.
