Speech Recognition Tech Can Transform Indian Agro Sector


Today, speech recognition is considered as one of the disruptive technologies that makes huge impact in banking, insurance, retail and manufacturing industries. The global market value of speech analytics is estimated to grow USD $456 million in 2014 to 1.33 billion by the year 2019 according to reports from Markets and Markets.

Speech recognition solutions enables natural, human-like conversations with machines and can achieve very high accuracy levels with appropriate fine tuning around vernacular attributes.By its enormous growth and rising popularity, it is becoming increasingly mainstream among different industries and enterprises. In India, where 70% of the country’s population is involved in the agriculture industry, speech technology has started been playing a critical role through user friendly speech solutions to rural farmers.

Disrupting agriculture industry in India with newer technology such as speech recognition is no easy feat. As farmers are wide-spread across different parts of the country and speak a variety of local languages, approaching them with a generic solution will not add value to them. Also the solutions should vary in functionality and platform from mainstream mobility solutions, as this industry and its stakeholders primarily use basic feature phones, and their literacy level is low. Here is where, huge benefits can be gleaned out of speech recognition solutions by the Indian agriculture industry.

Broadly speaking, there are three major functions in which agriculture sector is leveraging speech recognition technologies:

Providing access to information: Speech recognition is now being leveraged by leading Agribusinesses to provide famers with access to useful, relevant information. The speech recognition engine converts the voice query to text, identifies the appropriate answer from the enterprise databank, and provides the appropriate answer to the caller.  If the answer is not readily available, the query is recorded for an expert’s review, whose answer will then be played to the farmer through an outbound call.As an example, a farmer can dial a number, ask and get commodity prices in various urban markets using a simple featurephone. Asymmetrical access to pricing information is a weakness of rural markets. These applications are now being adoptedby governments and nonprofits to rule-out the above limitations and have the crucial information flow to farmers across India.

Obtaining feedback on various products and services: Many agribusinesses and governmentdepartments are using speech recognition technologies to obtain feedback from farmers on their products and services. Using these applications, automated calls are triggered to ask questions through human-like conversations to obtain famers’ voice feedback. Thus farmers’ voice response to the questions are recognized and converted into text data to create reports.

Automating field force support activity: Typically, Agribusinesses employ a network of human agents tocontinually visit farmers and provideinformation and various services to them. These field agents serve as the primary communication channel between farmers and agribusinesses in the country. They are responsible for capturing on-field information using pen-and-paper forms and pass it to the back office in regular intervals. This manual pen-and-paper process causes delay in field operations and not transferring vital information to stakeholders on time. Using speech recognition applications, field agents’ are now transferring information to back office in real-time through their mobile devices. This significantly reduces travel costs as well as expenses associated with pen and paper.

Together, these three types of applications using speech recognition technologies enable more efficient operations across agriculture industry. For instance, Uniphore has implemented a similar multi-lingual speech recognition solution for a state government in India to engage over 1.8 million farmers across the state.  First, Uniphore setup an outbound module for alerts, remindersand surveys.  Second, an inbound module was setup to record farmer query, notify expert(s) through SMS, record the expert’s reply and play it back to the farmer. Overall, Uniphore’s speech solutions helped the state government achieve following results:

Cost savings: The state department has cut down costs on human resources and different inefficient channels that were used to reach farmers. Cost of training and deploying both the field agents and call center agents were drastically reduced since most of the frequently asked questions and scenarios were automated and covered through speech solutions.

Higher engagement: On an average 70% of the farmers’ population were able to interact with the solution in a seamless way.  This higher engagement can be attributed to the fact that the solution covered native languages and thus enabled farmers to converse in their own vernacular.

Speech recognition technologies, in addition to bringing down operational costs, also support the livelihoods of farmers through a simple and convenient solution, creating a win-win scenario across stakeholders.