Automatic Speech Recognition
A technology that automatically converts human speech into text. It is a core subfield of speech processing and natural language processing, with broad applications in voice assistants, subtitle generation, and speech assessment.
Timeline (last 90 days)
The technical difficulty of transcript-driven editing lies in timestamp alignment precision: word-level timestamps output by ASR may have errors of tens to hundreds of milliseconds.
Homophone confusion is a classic hard problem in automatic speech recognition (ASR) systems; words with similar phonemes such as spell and smell can hardly be distinguished by sound alone when semantic or task context is lacking.
Mainstream ASR systems typically require thousands to tens of thousands of hours of human-annotated audio to achieve usable accuracy.
Semi-supervised federated learning (SSFL) trains an initial model using a small amount of labeled seed data on the server side, then uses a teacher model to generate pseudo-labels for large volumes of unlabeled client data, thereby optimizing the model without uploading raw audio.
The output of speech recognition is a sequence, and errors in pseudo-labels accumulate across the entire output sequence, where a recognition deviation at one position may trigger a chain of misjudgments in subsequent tokens.
In ASR scenarios, the teacher model is typically an end-to-end speech recognition model based on CTC or an Attention-based Encoder-Decoder.
An arXiv paper (arXiv:2609.12122) proposes using acoustic masking combined with automatic speech recognition (ASR) models to automatically quantify each consonant's contribution to word intelligibility.
The research method works by silencing one consonant in a word at a time and testing whether the ASR model can still correctly recognize that word.
MMR measures whether a machine can still recognize the word rather than whether a human listener can still understand it, and there are systematic differences between ASR model recognition behavior and human perception.
AI real-time caption translation typically involves two layers of technology: an ASR model first converts speech into source-language text, which is then translated into the target language by an NMT model.
2 more timeline events
All Facts (12)
The technical difficulty of transcript-driven editing lies in timestamp alignment precision: word-level timestamps output by ASR may have errors of tens to hundreds of milliseconds.
50%UnverifiedHomophone confusion is a classic hard problem in automatic speech recognition (ASR) systems; words with similar phonemes such as spell and smell can hardly be distinguished by sound alone when semantic or task context is lacking.
50%UnverifiedMainstream ASR systems typically require thousands to tens of thousands of hours of human-annotated audio to achieve usable accuracy.
50%UnverifiedSemi-supervised federated learning (SSFL) trains an initial model using a small amount of labeled seed data on the server side, then uses a teacher model to generate pseudo-labels for large volumes of unlabeled client data, thereby optimizing the model without uploading raw audio.
50%UnverifiedThe output of speech recognition is a sequence, and errors in pseudo-labels accumulate across the entire output sequence, where a recognition deviation at one position may trigger a chain of misjudgments in subsequent tokens.
50%UnverifiedIn ASR scenarios, the teacher model is typically an end-to-end speech recognition model based on CTC or an Attention-based Encoder-Decoder.
50%UnverifiedAn arXiv paper (arXiv:2609.12122) proposes using acoustic masking combined with automatic speech recognition (ASR) models to automatically quantify each consonant's contribution to word intelligibility.
50%UnverifiedThe research method works by silencing one consonant in a word at a time and testing whether the ASR model can still correctly recognize that word.
50%UnverifiedMMR measures whether a machine can still recognize the word rather than whether a human listener can still understand it, and there are systematic differences between ASR model recognition behavior and human perception.
50%UnverifiedAI real-time caption translation typically involves two layers of technology: an ASR model first converts speech into source-language text, which is then translated into the target language by an NMT model.
50%UnverifiedAutomatic speech recognition can still make errors in scenarios involving technical terminology, multiple speakers, and accents, directly affecting retrieval quality.
50%UnverifiedYouTube's automatic speech recognition (ASR) system can generate timestamp-aligned captions for the vast majority of videos, providing fine-grained temporal alignment between video frames and speech text.
50%