Symbl.ai unlocks access to state of the art large language models for all types of communication data. It enables enterprises to analyze multi-party conversations and transform unstructured conversations into knowledge, events, and insights.

There are three core components of the platform:

1. Automatic Speech Recognition: By default, using our ASR, customers can convert audio and video conversations to text in real-time and asynchronously. Real time speech recognition is supported with WebSocket and telephony protocols such as SIP and PSTN. Asynchronous speech recognition is supported via REST interface. Customers can also submit transcripts generated with their own ASR using our Async Text API.

2. Real time AI: After processing a conversation, customers can use APIs such as Entities, Trackers, Action Items, Sentiments and more for extracting task specific conversation intelligence in real time.

3. Generative AI: The platform enables generative capabilities through Nebula LLM - a foundation model for conversation data and performing instructed tasks in the context of the conversation such as summarization, information extraction, report generation, and others. The Nebula LLM is augmented by Nebula embeddings - an embeddings model using which customers can transform conversations into vectors and feed as a context to the Nebula LLM to perform generative tasks using historical data.