#SpeechScience
📚 Citation Classic
"SpecAugment: A simple data augmentation method for automatic speech recognition"
Park et al (2019)
Citations: 3,950+
A deceptively simp...
🔗 https://scholar.google.com/scholar?q=Park+2019+SpecAugment+simple+data+augmentation+method+automatic+speech+recognition
#SpeechScience
November 19, 2026 at 7:27 PM
Alvin Liberman proposed (1967) that we perceive speech by mentally simulating how we would produce it. This controversial theory sparked decades of debate about the link between perception and production.

🧠 Liberman et al. - Motor Theory paper

#SpeechScience #Theory
November 15, 2026 at 4:57 PM
A small patch in the brain’s listening region carries clues to your next word about a quarter of a second before you say it. The finding offers a possible target for devices that help people regain speech.

#NeuroSky #SpeechScience
intouchwithscience.com/story/brain-...
A listening region of the brain helps prepare your next word
The discovery offers a possible target for devices that help people communicate after losing speech.
intouchwithscience.com
November 20, 2026 at 1:01 AM
📚 Citation Classic
"Signal estimation from modified short-time Fourier transform"
Griffin & Lim (1984)
Citations: 4,150+
The Griffin-Lim algorithm for recoverin...
🔗 https://scholar.google.com/scholar?q=Griffin+Lim+1984+Signal+estimation+from+modified+short-time+Fourier+transform
#SpeechScience
November 12, 2026 at 6:48 PM
In 1939, Bell Labs unveiled the Voder at the New York World's Fair - the first electronic speech synthesizer operated by a human using a keyboard and foot pedals. Audiences were amazed to hear a machine 'speak'!

🎪 See it in action: youtube.com/watch?v=0rAyrmm7vv0

#SpeechScience #Technology
November 8, 2026 at 4:38 PM
📚 Citation Classic
"Speech recognition with deep recurrent neural networks"
Graves, Mohamed & Hinton (2013)
Citations: 7,350+
Showed deep LSTMs could push p...
🔗 https://scholar.google.com/scholar?q=Graves+Mohamed+Hinton+2013+Speech+recognition+with+deep+recurrent+neural+networks
#SpeechScience
November 5, 2026 at 6:38 PM
In the 1960s researchers found speech sounds overlap - when saying 'key' your lips round for the vowel while still pronouncing the 'k'. This is why isolated phonemes don't exist in natural speech.

🔄 Sounds blend continuously

#SpeechScience #Phonetics
November 1, 2026 at 4:45 PM
📚 Citation Classic
"Phoneme recognition using time-delay neural networks"
Waibel et al (1989)
Citations: 4700+
Time-delay neural networks brought early deep learning to phon...
🔗 https://scholar.google.com/scholar?q=Waibel+1989+Phoneme+recognition+using+time-delay+neural+networks
#SpeechScience
October 29, 2026 at 5:23 PM
In the 1960s-70s, LPC revolutionized speech processing by modeling the vocal tract as a linear filter. This efficient compression method enabled digital speech transmission and became the basis for early voice calls.

📞 Foundation of digital telephony

#SpeechScience #Technology
October 25, 2026 at 2:16 PM
📚 Citation Classic
"Some experiments on the recognition of speech with one and with two ears"
Cherry (1953)
Citations: 5600+
Origin of the 'cocktail party problem' — how we tune...
🔗 https://scholar.google.com/scholar?q=Cherry+1953+recognition+of+speech+with+one+and+with+two+ears
#SpeechScience
October 22, 2026 at 3:53 PM
In the 1930s, Stevens and Volkmann showed that our perception of pitch isn't linear - the mel scale was born! This perceptual scale is now fundamental to speech processing and voice technology.

🎵 Foundation of modern speech recognition systems

#SpeechScience #Perception
October 18, 2026 at 1:55 PM
📚 Citation Classic
"Second language speech learning: Theory findings and problems"
Flege (1995)
Citations: 5700+
Speech Learning Model explaining why L2 sounds are so...
🔗 https://scholar.google.com/scholar?q=Flege+1995+Second+language+speech+learning+theory+findings+and+problems
#SpeechScience
October 15, 2026 at 3:35 PM
During WWII, engineers at Bell Labs developed the sound spectrograph to analyze encrypted speech. After the war, it became the speech scientist's most important tool for visualizing sound.

🎙️ Learn more: antiqueradio.org/soundspec.htm

#SpeechScience #Technology
October 11, 2026 at 4:14 PM
📚 Citation Classic
"The Sounds of the World's Languages"
Ladefoged & Maddieson (1996)
Citations: 6600+
Encyclopedic survey of consonant articulations across the world's langua...
🔗 https://scholar.google.com/scholar?q=Ladefoged+Maddieson+1996+The+Sounds+of+the+World%27s+Languages
#SpeechScience
October 8, 2026 at 5:56 PM
Your voice may reveal how fast you're ageing. A new speech age gap predicts brain, biological & cognitive decline. #Ageing #BrainHealth #AI #SpeechScience #Dementia
freegardner.com/sci_scout/sp...
Speech Age Gap Predicts Accelerated Ageing
The researchers were not hunting for a way to measure ageing. They were building a machine learning system to analyse speech hundreds of acoustic and...
freegardner.com
October 5, 2026 at 6:40 PM
Gunnar Fant's source-filter theory (1960) explained how speech is produced: a sound source (vocal folds) filtered by the vocal tract shape. This model is still the foundation of speech science today!

📖 Reference: Fant's Acoustic Theory of Speech Production

#SpeechScience #Theory
October 4, 2026 at 3:15 PM
📚 Citation Classic
"Conversational interfaces: Advances and challenges"
Zue & Glass (2000)
Citations: 720+
Surveyed conversational speech interfaces and the challenges of sp...
🔗 https://scholar.google.com/scholar?q=Zue+Glass+2000+Conversational+interfaces+advances+and+challenges
#SpeechScience
October 1, 2026 at 5:31 PM
In 1779, Christian Kratzenstein built one of the first speech synthesizers - a set of resonating tubes that could produce vowel sounds when air was blown through them.

🔗 More: en.wikipedia.org/wiki/Speech_synthesis#History

#SpeechScience #Technology
September 27, 2026 at 3:06 PM
📚 Citation Classic
"Word association norms mutual information and lexicography"
Church & Hanks (1990)
Citations: 7600+
Popularized mutual information for finding...
🔗 https://scholar.google.com/scholar?q=Church+Hanks+1990+Word+association+norms+mutual+information+and+lexicography
#SpeechScience
September 24, 2026 at 3:47 PM
During WWII, engineers at Bell Labs developed the sound spectrograph to analyze encrypted speech. After the war, it became the speech scientist's most important tool for visualizing sound.

🎙️ Learn more: antiqueradio.org/soundspec.htm

#SpeechScience #Technology
September 20, 2026 at 2:06 PM
📚 Citation Classic
"Robust text-independent speaker identification using Gaussian mixture speaker models"
Reynolds & Rose (1995)
Citations: 7600+
GMM-based...
🔗 https://scholar.google.com/scholar?q=Reynolds+Rose+1995+Robust+text-independent+speaker+identification+Gaussian+mixture
#SpeechScience
September 17, 2026 at 3:32 PM
CONGRATULATIONS to Dr. Rion Iwasaki who successfully defended his dissertation! Rion examined tongue movement during so-called devoiced high vowels in Tokyo Japanese. The study proves devoiced /i/ and /u/ are contrastive in both acoustic and articulatory domains.

#speechscience #CUNYGC #phdlife
September 15, 2026 at 4:18 PM
In the 1950s, researchers discovered that vowels could be characterized by their formant frequencies https://newt.phys.unsw.edu.au/jw/formant.html.

📚 Key paper: Potter et al. (1947) 'Visible Speech'

#SpeechScience #Discovery
September 13, 2026 at 2:15 PM
📚 Citation Classic
"WaveNet: A Generative Model for Raw Audio"
van den Oord et al (2016)
Citations: 7900+
WaveNet generated raw audio sample-by-sample — a leap for speech synthesis quality.
🔗 https://arxiv.org/abs/1609.03499
#SpeechScience
September 10, 2026 at 2:59 PM
In 1939, Bell Labs unveiled the Voder at the New York World's Fair - the first electronic speech synthesizer operated by a human using a keyboard and foot pedals. Audiences were amazed to hear a machine 'speak'!

🎪 See it in action: youtube.com/watch?v=0rAyrmm7vv0

#SpeechScience #Technology
September 6, 2026 at 1:31 PM