Analyzing The Quality And Stability Of A Streaming End-to-end On-device Speech Recognizer

Abstract

The demand for fast and accurate incremental speech recognition increases as the applications of automatic speech recognition (ASR) proliferate. Incremental speech recognizers output chunks of partially recognized words while the user is still talking. Partial results can be revised before the ASR finalizes its hypothesis, causing instability issues. We analyze the quality and stability of on-device streaming end-to-end (E2E) ASR models. We first introduce a novel set of metrics that quantify the instability at word and segment levels. We study the impact of several model training techniques that improve E2E model qualities but degrade model stability. We categorize the causes of instability and explore various solutions to mitigate them in a streaming E2E ASR system. Index Terms: ASR, stability, end-to-end, text normalization,on-device, RNN-T

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyshangguan2020analyzing

Related papers

• Streaming End-to-end Speech Recognition For Mobile Devices (2018)18.87
• A Streaming On-device End-to-end Model Surpassing Server-side Conventional Model Quality And Latency (2020)15.00
• Large-scale Multilingual Speech Recognition With A Streaming End-to-end Model (2019)14.97
• Two-pass End-to-end Speech Recognition (2019)13.97
• Recognizing Long-form Speech Using Streaming End-to-end Models (2019)13.74
• A Review Of On-device Fully Neural End-to-end Automatic Speech Recognition Algorithms (2020)9.92
• Modality Confidence Aware Training For Robust End-to-end Spoken Language Understanding (2023)2.26
• Joint Optimization Of Streaming And Non-streaming Automatic Speech Recognition With Multi-decoder And Knowledge Distillation (2024)0.00