Sprinklr AI/AI+ Service Features
Updated
Sprinklr AI+ is an advanced AI-powered suite integrated into Sprinklr's Unified-CXM platform, designed to transform unstructured customer experience (CX) data into actionable services. The following table gives details on multiple AI/AI+ features of Sprinklr Service.
Module | Feature Name | Feature Description | KB Article Link | Available in AI+ Studio (Y/N) | Can feature be turned on or off? | Is this feature access controlled? |
Conversational AI | ASR | STT, or Speech To Text AI models are machine learning-based systems designed to transcribe spoken language into written text. Sprinklr uses STT Models extensively in Sprinklr Service to resolve customer queries with voice assistants, real time transcriptions or Speech Analytics solutions. | No | Yes | No | |
Conversational AI | Text to Speech | Text to Speech or TTS is a machine learning based language model that synthesizes text into a desired voice. The goal is to generate synthetic speech that sounds natural and resembles human speech as closely as possible. Now with Sprinklr’s all new TTS technology, give your customers a human-like experience while interacting with voice assistants. | No | Yes | No | |
Conversational AI | Intent Detection | Intent Detection is a foundational component that helps understand the main reasons behind each customer interaction. | No | No | No | |
Conversational AI | Intent Detection | Intent Detection is a foundational component that helps understand the main reasons behind each customer interaction. | No | No | No | |
Conversational AI | Entity Detection | Entity Detection is a foundational component that identifies and extracts specific pieces of information from each customer or agent utterance. By detecting relevant data—such as products, locations, or account details—bots can personalize customer journeys, enhance resolution speed, and provide more accurate assistance. | No | No | No | |
Conversational AI | Entity Detection | Entity Detection is a foundational component that identifies and extracts specific pieces of information from each customer or agent utterance. By detecting relevant data—such as products, locations, or account details—bots can personalize customer journeys, enhance resolution speed, and provide more accurate assistance. | No | No | No | |
Conversational AI | Discovery Run | Brands can enhance their bot's quality and efficiency by discovering contact drivers for bot workflows through a LLM powered data-driven approach. This eliminates the need for guesswork when implementing bots, while ensuring the coverage for customer journeys. | No | No | No | |
Conversational AI | Dynamic Workflow | No | Yes | No | ||
Conversational AI | Dynamic Workflow | No | Yes | No | ||
Conversational AI | FAQ Bots | FAQ chatbots are RAG-driven virtual assistants that automate responses to common customer inquiries, providing instant, accurate answers. They utilize NLP and RAG to understand user queries and retrieve relevant information from a pre-built knowledge base. By handling routine questions, these chatbots free up human agents to focus on more complex issues, improving overall efficiency and customer satisfaction. FAQ bot leverages Large Language Models (LLMs) to interpret and respond to a broader range of user questions, including those not explicitly covered in the predefined knowledge base. This capability enables them to generate contextually relevant answers, enhancing the user experience and providing more comprehensive support. | No | Yes | No | |
Conversational AI | FAQ Bots | FAQ chatbots are RAG-driven virtual assistants that automate responses to common customer inquiries, providing instant, accurate answers. They utilize NLP and RAG to understand user queries and retrieve relevant information from a pre-built knowledge base. By handling routine questions, these chatbots free up human agents to focus on more complex issues, improving overall efficiency and customer satisfaction. FAQ bot leverages Large Language Models (LLMs) to interpret and respond to a broader range of user questions, including those not explicitly covered in the predefined knowledge base. This capability enables them to generate contextually relevant answers, enhancing the user experience and providing more comprehensive support. | No | Yes | No | |
Outbound Voice | Answering Machine Detection | The Answering Machine Detection (AMD) system is an advanced ensemble of models designed to detect voicemails accurately in outbound calling campaigns. By using multiple models that work together, the AMD system improves detection accuracy, ensuring agents connect with live customers more effectively. This results in higher productivity for agents and better customer engagement, as fewer calls are misrouted to voicemails, allowing agents to focus on live interactions. | No | Yes | UI might not be access controlled, but access can be controlled from backend. | |
Outbound Voice | Predictive Dialer | Sprinklr’s Predictive Dialer utilizes a Reinforcement Learning (RL) algorithm to optimize outbound voice dialer campaigns. Unique to Sprinklr, this RL-based approach dynamically adjusts pacing based on real-time environmental factors, ensuring efficient call distribution and minimizing idle time, while maximizing agent productivity and customer engagement. | No | Yes | UI might not be access controlled, but access can be controlled from backend. | |
Agent Assist | Smart Comprehend | Get proactive KB/FAQ articles suggestions based on customer care conversations on live cases. | No | Yes | No | |
Agent Assist | Smart Responses | Smart Response employs RAG to learn from past agent-customer conversations and articulate relevant response suggestions. | No | Yes | Yes | |
Agent Assist | Smart Compose | Smart Compose accelerates crafting of replies by delivering AI-backed phrase completion for agents. It uniquely does this by learning from past conversations to offer personalized suggestions, which can be swiftly selected with a Tab or Right Arrow Key. | No | Yes | No | |
Agent Assist | Similar Cases | Sprinklr’s AI identifies and matches the current customer’s issues or queries with contextually similar resolved cases, providing agents with valuable insights and historical resolutions. This guidance helps agents respond quickly and effectively, improving overall case resolution efficiency. | No | Yes | Yes | |
Agent Assist | Smart Paraphraser | Smart Paraphraser enables agents to send articulate, grammatically correct responses faster with AI-powered paraphrased suggestions. Paraphrasing will help agents to bring much-needed variety to conversations as it will help in reducing the robotic nature of conversations by enlarging the scope of usual agent replies. | No | Yes | Yes | |
Agent Assist | Agent Nudges | No | Yes | Yes | ||
Agent Assist | Smart Assist Search | No | ||||
Agent Assist | Predictive CSAT | Predicted CSAT is ML regression problem statement, it is a scoring mechanism that quantifies customer happiness using AI on a scale of 1-100. | No | No | No | |
Conversational Analytics | Case Summary | Get a Generative AI driven case summary of the conversation between a customer and the brand upon request without having to read the full conversation. | No | Yes, disable the rule & edit record page. | No | |
Conversational Analytics | Predictive CSAT | Predicted CSAT is ML regression problem statement, it is a scoring mechanism that quantifies customer happiness using AI on a scale of 1-100. | No | No | No | |
Conversational Analytics | Contact Driver Discovery | This generative AI based approach is responsible for uncovering new or previously unidentified customer intents through unsupervised or semi-supervised methods. By analyzing conversational data, it identifies recurring themes or questions that can be classified as new intents. It is designed to interact with customer-agent conversations, leveraging AI to enhance business intelligence by identifying, categorizing, and managing intents within conversations. | No | Yes | Yes | |
Conversational Analytics | Voice Acoustics | Voice acoustics play a crucial role in shaping the quality of communication during voice calls, especially in environments like customer support. Key elements of voice acoustics include Signal-to-noise ratio (SNR), Loudness, and Rate of Speech (RoS), each of which can significantly affect the clarity, comprehension, and overall effectiveness of a conversation. Analyzing these acoustic features helps improve the quality of calls, enhances user experience, and provides valuable insights for speech applications. | No | No | Yes | |
Conversational Analytics | Statistical Stories | Given a large dataset of customer-agent conversations, its crucial to extract temporal anomalies and/or trends for similar kind of conversations. Using “Statistical Insights”, we generate anomalous collection of conversation cases with similar dimensions and/or conversation intents, which have an unexpectedly positive or negative trend in the latest day/week/month. This proves crucial in analysing latest trends of the client’s product/service performance. | No | Yes, update enabled insight type via red. | Yes | |
Conversational Analytics | Agent Stories | Given a large dataset of customer-agent conversations, its crucial to extract granular and focused information for agents employed by partners. Using “Agent Focused Insights”, we generate actionable stories, which take into consideration conversations from similar agent focused issue root causes and agent skills. | No | Yes, update enabled insight type via red. | Yes | |
Conversational Analytics | Product Stories | Given a large dataset of customer-agent conversations, its crucial to extract granular and focused information for Product/Service based partners. Using “Product or Service Focused Insights”, we generate actionable stories, which take into consideration conversations from similar product/service focused issue root causes, focus areas and conversational intents. | No | Yes, update enabled insight type via red. | Yes | |
Conversational Analytics | Contact Driver Gap Insights | Given a taxonomy and case conversations belonging to a particular brand, this feature suggests new Contact Driver Intents that could be added to the existing taxonomy of any number of levels. These new intents are generated on the basis of the case conversations that could not be aptly matched to any of the existing intents. | No | Yes, update enabled insight type via red. | Yes | |
Conversational Analytics | KB Gap Analysis | KB Gap Analysis aims at improving KB articles by adding information to existing KBs or creating new KB Articles by using information present in customer-brand conversations. | No | Yes, update enabled insight type via red. | Yes | |
Conversational Analytics | KB Refinement | This feature analyses any number of structured or unstructured knowledge bases (KBs) to identify redundant and ambiguous information. It provides recommendations comprising of the following two sections: Suggesting multiple KBs that could be merged to reduce duplication and scattered data. Suggesting potentially ambiguous information across KBs. | No | Yes | Yes | |
Knowledge Base | KB Gap Analysis | KB Gap Analysis aims at improving KB articles by adding information to existing KBs or creating new KB Articles by using information present in customer-brand conversations. | No | Yes | No | |
Knowledge Base | KB Refinement | This feature analyses any number of structured or unstructured knowledge bases (KBs) to identify redundant and ambiguous information. It provides recommendations comprising of the following two sections: Suggesting multiple KBs that could be merged to reduce duplication and scattered data. Suggesting potentially ambiguous information across KBs. | No | Yes | No | |
Workforce Management | Scheduling | Generating optimal scheduling solutions that allocate agents into specific days and shifts while adhering to various business constraints. This scheduling system aims to minimize downtime and operational disruptions by translating business requirements into quantifiable constraints and we utilize the most effective objective function to achieve the best possible scheduling outcomes. | No | No, but we can control the visibility of the module in the persona of a user. This feature is a part of basic WFM enablement step. | Yes | |
Agent Assist | Case Summary | Get a Generative AI driven case summary of the conversation between a customer and the brand upon request without having to read the full conversation. | Yes | Yes | No | |
Agent Assist | Writing Assistance | Writing Assistance helps you generate relevant and engaging content for posts, reels, and messages, all within the Quick Publisher. It also provides proofreading and editing tools—including grammar and spelling checks—to ensure high-quality content creation across multiple channels. Users can quickly reword, simplify, or expand existing text, modify tone, translate content, and even generate hashtags or tailored post variations. | Yes | Yes | Yes | |
Conversational Analytics | Top Contact Drivers | Understanding the key drivers of customer interactions and gaining valuable insights into complex business inquiries can be a challenging task. With the help of AI Contact Driver analysis, it is possible to analyze 100% of customer-agent conversations to identify the primary reasons why customers reach out to your brand. This approach can uncover opportunities to streamline your customer service and eliminate unnecessary contacts. | Yes | Yes, disable the rule & edit record page OR disable Contact Drivers DP. | Yes | |
Conversational Analytics | AQM | AQM is a NLP tool designed to enhance agent performance and improve customer satisfaction by analyzing call transcripts and evaluating agent responses against a predefined set of questions. | Yes | Yes | Yes | |
Agent Assist | Sprinklr AI+ Field | Generate Case Summary via AI+. | No | No | No | |
Agent Assist | Groovy AI Generator | Generate Groovy scripts, summarize existing ones, debug issues, and modify scripts to achieve complex use cases. This reduces the time and effort involved in writing Groovy scripts, making it easier for users, even those not proficient in Groovy, to accomplish complex tasks. | Yes | Yes | Yes | |
Voice | AI Autofill | AI Autofill is an AI-powered feature designed to streamline and automate After-Call Work (ACW) in contact centers. It leverages artificial intelligence prefill call summaries in asw as disposition form ,reducing agent workload and improving data quality. | Yes, using a DP. | Indirectly. We can control the access of ACW where this feature is enabled. |