Cloud Telephony Experiments

Ever since WhatsApp launched their business API last year, businesses have been waiting to enable the WhatsApp support channel. WhatsApp has slowly been rolling out API support to select businesses. Though most use cases have been around chat bots, there is a good use case for human agent support on the WhatsApp channel. We at Ozonetel were one of the first companies which enabled agent support for the WhatsApp channel as soon as the business API released. You can read more about it at https://blog.kookoo.in/2018/08/kookoo-contact-center-integration-with.html. Today we have improved the integration by connecting with Zendesk. So if you are serving your customers using Zendesk, you can easily enable WhatsApp channel for your customers. Your customers can chat with your agents directly on WhatsApp. We already have one of the best CTI integrations with Zendesk which you can install from the Zendesk marketplace, https://www.zendesk.com/apps/support/kookoo/ The same app can now be ...

Well, it's finally here. The number masking API. To be fair, KOOKOO has supported number masking since when we launched the <dial> tag and already a lot of our customers use our number masking services. But this launch makes it easier for new developers to use the virtual number services. Instead of building a full KOOKOO app and hosting it, you can just call the API. Flickr What is it? Number mask API allows you to create conversation sessions between your customers and your mobile workforce without revealing their personal phone numbers. Create virtual phone numbers instantly with this API. Who can use this? Any business which needs to protect the privacy of its customers and it's employees will find this API useful. For example, ride hailing services a-la Uber, Ola etc can use this API to create virtual number sessions per ride. Delivery services like Zomato, Foodpanda, Swiggy etc can use this API to create virtual numbers on demand per delivery. How can...

WhatsApp announced the launch of their business API yesterday. This is a big deal. A lot of businesses have been asking for WhatsApp integration and now its a possibility. Twilio has already launched an integration . Won't be long before we see a lot more integrations pop up. The most obvious use case is for support chat bots, alerts and reminders. But we at Ozonetel believe one other use case will become the dominant use case, WhatsApp as a support channel with human agents. We believe WhatsApp will become a dominant channel in contact centers world wide. There are multiple reasons for this: WhatsApp is ubiquitous. Everyone has the app. WhatsApp interface is well known to all customers. It is feature rich. You can share audio, video, location etc in the app itself. You can integrate bots for repetitive tasks. Being the leaders in cloud contact center solutions, we at Ozonetel knew that we had to integrate with the WhatsApp channel as early as possible....

TLDR: The data sets used for the testing in the experiments are the digit examples from the Google speech commands dataset . Experiment 1: Converted dataset to 8Khz to fit the telephony format. Google API accuracy: 74.3% Kookoo Zena accuracy: 92.4% We felt the accuracy for Google dropped down because of the 8Khz format. So we did another experiment for Google speech API. Experiment 2: Original speech command dataset Google API accuracy: 85.6% The Long version: At Ozonetel, we are constantly innovating and we are back again with a new speech model after the Yes/No model. We all saw Google demo Duplex last week and it was a pretty cool demo. Just to put things in perspective, though the demo was cool, we think we are still some a ways away from actually having a system operate as smoothly as shown in the demo. We come to the conclusion mainly based on our experiments with the Google speech API. Unless Google is using some model other than the one exposed publicly ...

Unless you have been living under a rock, you know by now that Google announced Google Duplex and took the world over by storm. One of the main building blocks of the Google Duplex system is its natural sounding TTS(text to speech) engine. Though, Google has not opened up those exact voices as a service, they have enabled a lot of great sounding text to speech voices as part of their cloud speech service . And as of today, KooKoo supports the Google TTS engine. You can now use Google TTS voices to power your IVR flows and make them more natural sounding. Adding Google TTS to your IVR is as simple as just adding an attribute type="ggl" in your <playtext> tags . And the best news of all, this service is free on KooKoo till August 2018. So go ahead, replace your current TTS voices in your IVR flows with Google voices and see the difference. Google TTS has lots of options. In the coming weeks, we will be opening up those options through the KooKoo APIs. So keep w...

A couple of months back, we had launched the Kookoo <recognize> tag. And as mentioned in the blog post, we support the Google speech engine and our very own Zena engine. The Kookoo speech engine as it currently stands is just a drop in the ocean when compared to the features provided by the Google speech engine. But, as they say, a journey of a thousand miles starts with a single step. This post is about our first step. As of today, we are launching our first model, the Yes/No model, built in house, based on a proprietary AI algorithm . You can now design your Kookoo IVRs to say "Please say yes or no", instead of "Please press 1 for yes and 2 for no". In the internal tests our model's accuracy has been so good, we believe that this can become the standard interaction mechanism instead of DTMF. We also did a side by side comparison of our model with Google's speech API. Before presenting the results below, some disclaimers: 1. Our model is a ...

Just recently we launched the Kookoo speech API . This will allow your applications to hear what your customers are saying. We also showed you how to convert your chatbots to voicebots . So whats next? By now, many of you would have realized that we are going for a full stack approach. We started off with the telephony stack and then added a speech layer on top of it. Today, we are announcing the addition of an understanding layer, Kookoo Intent (beta). This is an invite only launch, so please send us an email if you are interested to try this out. To be fair, you can just use our telephony stack and the speech layer and then use some third party APIs for intent analysis. There are a lot of APIs for Intent from Google, Facebook etc. But if you just want to build your complete bot using Kookoo, then Kookoo Intent allows you to do that. More importantly, we will be concentrating on providing NLP services for Indian languages. We will be partnering with NLP API providers to pow...