Slang labs is a Bengaluru-based startup that has launched a full-stack solution that provides smart and accurate multilingual voice search capabilities within e-commerce apps known as Conva.
The startup is backed by google and the solution is designed to facilitate seamless integration into existing e-commerce apps in under 30 minutes without requiring developers to have any knowledge of advanced voice tech stack concepts such as Automatic Speech Recognition (ASR), Natural Language Processing (NLP), and Text-to-Speech (TTS).
CONVA’s voice search technology is capable of comprehending mixed-code utterances, allowing users to speak naturally in their own language while searching for products and information within e-commerce mobile and web apps. Brands can continue to maintain their app backend in a single language, typically English, while still offering multilingual voice search capabilities to their users.
The solution is available as a simple Software Development Kit (SDK) that can be easily integrated into existing e-commerce apps. CONVA’s smartness in handling multiple variations of product names and filters across multiple languages makes it a much more powerful voice search experience for apps than any other product in the world, according to Kumar Rangarajan, Co-founder of Slang Labs.
The technology behind CONVA has been trained on lakhs of SKUs from various industries, including groceries, FMCG, medicines, fashion, beauty products, cosmetics, and food names. The models have also been pre-trained on thousands of Indian place names, train stations, airport terminals, bus pickup and drop-off points inside of cities and towns, as well as all the equities traded on the BSE and NSE exchanges.
CONVA’s precision has been evaluated and determined to be up to 46% more precise than Google ASR in regard to voice search. This level of accuracy is a significant factor in why some of India’s most prominent e-commerce applications trust CONVA.
Currently, CONVA is accessible for implementation in e-commerce, retail, travel, and trading applications. The solution’s features can assist businesses in offering a more immersive and customized user experience by enabling users to look for products utilizing their natural language, without the need to switch languages.
CONVA-powered voice search offers a seamless search experience to customers, allowing them to conveniently search for products within the application using their typical colloquial terms, even when mixing languages within a sentence. The apps can still accurately recognize the desired product being searched.
Kumar Rangarajan, one of the Co-founders of Slang Labs, has introduced CONVA, which is an acronym for COnversational iN-app Voice Assistant. CONVA offers an advanced voice search experience for apps, surpassing any other product in the world due to its ability to handle multiple variations of product names and filters in various languages.
According to Rangarajan, CONVA has made significant technological progress, resulting in a voice search accuracy level that surpasses Google ASR by up to 46%. This is why leading Indian e-commerce apps have confidence in CONVA.
CONVA is presently accessible for utilization in e-commerce, retail, travel, and trading apps. Its models have been trained on lakhs of Stock Keeping Units (SKUs) from a wide range of industries, such as groceries, fast-moving consumer goods (FMCG), pharmaceuticals, fashion, beauty items, cosmetics, and food items.
CONVA models are also pre-trained on thousands of Indian place names, 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.
