On-Device Speech Recognition
A speech recognition approach in which the acoustic model and language model are deployed to run on the user's device. Audio does not leave the local environment, there is no network round-trip latency, and the system can operate offline, but it is constrained by model capacity.
Timeline (last 90 days)
On-device speech recognition typically relies on lightweight models that have been quantized and pruned, such as end-to-end models based on CTC or RNN-T architectures.
The recognition accuracy of on-device models is typically lower than that of large cloud models, especially in noisy environments or scenarios with many technical terms.
All Facts (2)
On-device speech recognition typically relies on lightweight models that have been quantized and pruned, such as end-to-end models based on CTC or RNN-T architectures.
50%UnverifiedThe recognition accuracy of on-device models is typically lower than that of large cloud models, especially in noisy environments or scenarios with many technical terms.
50%