Recently, the speech community is seeing a significant trend of moving from deep neural
network based hybrid modeling to end-to-end (E2E) modeling for automatic speech …

Contextual biasing is an important and challenging task for end-to-end automatic speech
recognition (ASR) systems, which aims to achieve better recognition performance by biasing …

Neural text-to-speech (TTS) generally consists of cascaded architecture with separately
optimized acoustic model and vocoder, or end-to-end architecture with continuous mel …

We previously proposed contextual spelling correction (CSC) to correct the output of end-to-
end (E2E) automatic speech recognition (ASR) models with contextual information such as …

Contextual Spelling Correction (CSC) models are used to improve automatic speech
recognition (ASR) quality given userspecific context. Typically, context is modeled as a large …

Contextual spelling correction models are an alternative to shallow fusion to improve
automatic speech recognition (ASR) quality given user vocabulary. To deal with large user …

Contextual biasing enables speech recognizers to transcribe important phrases in the
speaker's context, such as contact names, even if they are rare in, or absent from, the …

Recent studies have revealed that grammatical error correction methods in the sequence-to-
sequence paradigm are vulnerable to adversarial attacks. Large Language Models (LLMs) …

为了研究拼音对检测和纠正语音识别文本错误的影响, 提出了一种基于中文语义−
音韵信息的文本校对模型. 定义了5 种拼音编码方法构建字符− 音韵嵌入向量 …

Hybrid and end-to-end (E2E) systems have their individual advantages, with different error
patterns in the speech recognition results. By jointly modeling audio and text, the E2E model …