Peer-reviewed research
Lombard effect compensation and noise suppression for noisy Lombard speech recognition
Sang-mun Chi and Yung-Hwan Oh, International Conference on Spoken Language Processing (ICSLP) 1996, ISCA Archive, 1996
Read the originalRead on October 16, 2026
- What it is
- A 1996 conference paper that models how people change their speech when they talk in noise (the Lombard effect) and compensates for it in a recognizer. It reports that recognition degrades quickly once ambient noise is present.
- How much weight it carries
- Peer-reviewed conference paper, about 30 years old. Its tests used a small vocabulary of Korean words, so it shows the direction of the effect, not current accuracy in modern systems.
Facts taken from this source
- A 1996 study says that the performance of a speech recognition system degrades rapidly in the presence of ambient noise.
“The performance of speech recognition system degrades rapidly in the presence of ambient noise.” (Abstract)
Pages that rest on this source
- Dictation for RSI and Carpal Tunnel: Type Less by VoiceGuide
- How to Transcribe Audio to TextGuide
- Medical Dictation: Software, Devices and How It WorksGuide
- What Is Dictation Software? Types and How It WorksGuide
- iPhone Dictation Not Working? How to Fix ItGuide
- Voice Access in Windows 11: Setup and How It WorksGuide
- Speaking vs Typing Speed: What the Research ShowsGuide
- Dictation and Background Noise: Where and How to DictateGuide
- Can you dictate in a noisy room?Question
- Can you use dictation in an open-plan office?Question
- Can you use dictation or voice typing to transcribe a YouTube video?Question
- Can you whisper to dictation software?Question
- How do you transcribe audio that is playing on your computer?Question
- How do you transcribe low-quality or quiet audio?Question
- Is there a Whisper app for Windows?Question
- Why does dictating feel so awkward at first?Question
- Noise-canceling microphoneGlossary