Google-backed startup unveils multilingual voice search technology

Slang Labs, a Google-backed startup, said it has launched Conva, a full-stack solution that provides smart and highly accurate multilingual voice search capabilities inside e-commerce apps. 
 

The Bengaluru-based firm said that Conva is available as a simple SDK (Software Development Kit) that can be integrated into existing e-commerce apps in less than 30 minutes. Here developers don’t need any knowledge of Automatic Speech Recognition (ASR), Natural language processing (NLP), Text-to-Speech (TTS) and other advanced voice tech stack concepts. Conva stands for conversational in-app voice assistant.
 

“Conva’s smartness in handling multiple variations of product names and filters across multiple languages is what makes it a much more powerful voice search experience for apps than any other product in the world,” said Kumar Rangarajan, Co-founder of Slang Labs. “Several technological innovations in Conva make it up to 46 per cent more accurate than even Google ASR (automatic speech recognition) when it comes to voice search – which is why it is trusted by India’s most popular e-commerce apps.”
 

Slang Labs was co-founded in 2017 by founders who had earlier co-founded Little Eye Labs, the first and only Indian company to be acquired by Meta (Facebook). They had earlier worked in global companies like Apple, Microsoft Research, IBM and HP.
 

Conva-powered voice search comprehends mixed-code (multiple languages in one sentence) utterances. This enables consumers to speak naturally in their own language in order to search for products and information inside e-commerce mobile and web apps. This also allows the brand to maintain its app backend in only one language i.e. English. For instance, when people use English and another vernacular language within the same sentence for searching for something, Conva will understand both languages and provide a seamless search experience to the consumer. Customers can conveniently search for products inside the applications using their typical colloquial terms for well-known products using voice search that is enabled by Conva. And the apps will still be able to recognise the correct product being searched.
 

A few big players such as Amazon and Walmart-owned Flipkart have also built such technologies, but Slang Labs is focusing on the vast majority of organisations that lack such capabilities. Its technology acts as a layer on top of the existing apps of the brands.
 

 SlangLabs which has a team of 20 people in Bengaluru didn’t mention the companies that use Conva. But it said the technology is offered for use in e-commerce, retail, travel and trading apps. Its models have been trained on lakhs of SKUs (stock-keeping units) from a variety of industries, including groceries, FMCG, medicines, fashion, beauty products, cosmetics and food names. Conva models are additionally pre-trained on thousands of Indian place names. These include train stations, airport terminals, bus pickup and drop-off points inside cities and towns, as well as all the equities traded on the BSE and NSE exchanges.