Patent No. US11127636 (titled "Bot Group Messaging Using Bot-Specific Voice Libraries") was filed by Orion Labs Tech Llc on Mar 27, 2018.
’636 is related to the field of group messaging services and, more specifically, to enhancing such services with the use of software bots that can perform automated tasks. The background acknowledges the increasing prevalence of bots in various internet applications, including web crawling, information retrieval, and even communication via instant messaging. These bots can provide services like weather reports, currency conversion, and entertainment. The patent addresses the need for improved messaging services that allow users to interact with both user-oriented and group-oriented bots within group messaging environments.
The underlying idea behind ’636 is to enable seamless integration of bots into group messaging platforms by allowing the system to intelligently process audio messages and route them to the appropriate bot for action. This involves recognizing that a bot is part of a group, selecting a suitable voice library to convert the audio into a format the bot can understand (e.g., text), and then sending the processed message to the bot. The bot can then perform tasks based on the audio content, such as placing an order or providing information, and respond accordingly.
The claims of ’636 focus on a method, a memory device, and a computing system that facilitate the interaction between users and bots within a group messaging service. Specifically, the claims cover receiving a message with recorded audio and a bot identifier, searching a data structure to confirm the bot's membership in the group, accessing a bot entry to determine the appropriate voice library , selecting a voice library to process the audio, converting the audio into a format suitable for the bot, and sending the modified message to the bot.
In practice, the invention allows a user to send an audio message to a group, and if the message is intended for a bot within that group, the system automatically converts the audio into text using a selected voice library. This text is then sent to the bot, which can perform actions based on the message content. The selection of the voice library is crucial, as different bots may require different formats or processing methods. The system maintains a data structure that maps bots to specific voice libraries, ensuring that the audio is processed correctly for each bot.
This approach differs from prior solutions by providing a structured way to integrate bots into group messaging, including the intelligent selection of voice libraries for audio processing. Instead of simply forwarding audio to bots, the system ensures that the audio is converted into a format that the bot can understand and act upon. This allows for more complex interactions, such as placing orders or performing financial transactions, and enables the creation of more sophisticated and useful group messaging experiences. The use of shared bots and per-user bots further enhances the flexibility and customization of the system.
In the late 2010s when ’636 was filed, bot-based messaging platforms were gaining traction, at a time when speech-to-text conversion and natural language processing were increasingly integrated into these systems. Systems commonly relied on cloud-based services for processing audio and text, rather than on-device processing, when hardware or software constraints made real-time audio processing on mobile devices non-trivial.
The examiner approved the application because the independent claims (1, 10, and 17) teach a method, memory device, and computing system for bot messaging. The invention involves receiving a message with recorded audio and a bot identifier, searching a data structure to determine if the bot is a member of the group, accessing a bot entry with a voice library indicator, selecting a voice library to process the audio, processing the audio to produce a modified message, and sending the modified message to the bot member. The examiner stated that a single or obvious combination of prior art does not teach these limitations.
This patent contains 20 claims, with independent claims 1, 10, and 16. The independent claims are directed to a method, a memory device, and a computing system, respectively, all generally focused on processing audio messages within a group messaging service that includes bot members. The dependent claims generally elaborate on and refine the specifics of the method, memory device, and computing system described in the independent claims.
Definitions of key terms used in the patent claims.

The dossier documents provide a comprehensive record of the patent's prosecution history - including filings, correspondence, and decisions made by patent offices - and are crucial for understanding the patent's legal journey and any challenges it may have faced during examination.
Date
Description
Get instant alerts for new documents