Application-Agnostic Language Modeling for On-Device ASR
This addresses the problem of memory constraints in on-device ASR systems for applications like virtual assistants and speech-to-text, though it is incremental as it builds on existing language modeling approaches.
The paper tackled the challenge of building a single language model for on-device ASR that serves multiple applications without increasing memory, by proposing novel feed-forward architectures. The result was a model that reduces disk size by half while maintaining speed and accuracy compared to application-specific solutions.
On-device automatic speech recognition systems face several challenges compared to server-based systems. They have to meet stricter constraints in terms of speed, disk size and memory while maintaining the same accuracy. Often they have to serve several applications with different distributions at once, such as communicating with a virtual assistant and speech-to-text. The simplest solution to serve multiple applications is to build application-specific (language) models, but this leads to an increase in memory. Therefore, we explore different data- and architecture-driven language modeling approaches to build a single application-agnostic model. We propose two novel feed-forward architectures that find an optimal trade off between different on-device constraints. In comparison to the application-specific solution, one of our novel approaches reduces the disk size by half, while maintaining speed and accuracy of the original model.