Science Explorer Back to mapMap
Discussion
Topic · Artificial Intelligence

Speech Recognition and Synthesis

Speech Recognition and Synthesis is a research topic within Artificial Intelligence. Science Explorer counts 35k research works in it since 1950. 10.0% of them reached the world's top 10% most cited for their field and year.

This cluster of papers focuses on the advances in speech recognition technology, covering topics such as acoustic modeling using deep neural networks, speaker verification, convolutional neural networks for speech recognition, end-to-end speech recognition systems, hidden Markov models, sequence-to-sequence models, automatic speech recognition, speaker diarization, and statistical language modeling.

  • Deep Neural Networks
  • Acoustic Modeling
  • Speaker Verification
  • Convolutional Neural Networks
  • End-to-End Speech Recognition
  • Hidden Markov Models
  • Sequence-to-Sequence Models
  • Automatic Speech Recognition
  • Speaker Diarization
  • Statistical Language Modeling
Research works
35k
fractional, since 1950
In the world top 10%
3.6k
per year above
Top-10% rate
10.0%
share of its works in the world top 10%
Growth, 2018-21 → 2022-25
+58%
the tick is no change
Largest producer
China
35% of the 10 largest, 2022-now
Speech Recognition and Synthesis

Which countries lead Speech Recognition and Synthesis research?

By volume, China and India publish the most (2.4k and 1.3k works in 2022-now).

By volume, 2022-now

  1. 1 China 2.4k works
  2. 2 India 1.3k works
  3. 3 United States 1.2k works
  4. 4 Japan 472 works
  5. 5 United Kingdom 327 works
  6. 6 South Korea 321 works
  7. 7 Germany 246 works
  8. 8 France 205 works
  9. 9 Taiwan 190 works
  10. 10 Canada 150 works

How concentrated that is

The same countries as shares of everything the list above accounts for. A node where two countries do two thirds of the work and one spread evenly across twelve read alike as a ranking and not at all alike here.

China: 35.0%India: 19.4%United States: 17.5%Japan: 7.0%6 others listed: 21.2%35%largest
China2,373 · 35.0%India1,315 · 19.4%United States1,189 · 17.5%Japan472 · 7.0%6 others listed1,438 · 21.2%

Shares of the rows listed above, not of the whole node.

Which institutions lead Speech Recognition and Synthesis research?

By volume in 2022-now, Chinese Academy of Sciences publishes the most Speech Recognition and Synthesis research, followed by University of Science and Technology of China and Centre National de la Recherche Scientifique.

By volume, 2022-now

  1. 1 Chinese Academy of SciencesChina 94 works
  2. 2 University of Science and Technology of ChinaChina 73 works
  3. 3 Centre National de la Recherche ScientifiqueFrance 72 works
  4. 4 Alphabet (United States)United States 64 works
  5. 5 Shanghai Jiao Tong UniversityChina 63 works
  6. 6 Microsoft (United States)United States 63 works
  7. 7 Tsinghua UniversityChina 63 works
  8. 8 Xinjiang UniversityChina 55 works
  9. 9 Carnegie Mellon UniversityUnited States 54 works
  10. 10 Northwestern Polytechnical UniversityChina 45 works

Who are the leading researchers in Speech Recognition and Synthesis?

The most-cited researchers publishing on Speech Recognition and Synthesis include Yoshua Bengio, Geoffrey E. Hinton and Ross Girshick.

  1. 1 Yoshua Bengio Canada 476k citations
  2. 2 Geoffrey E. Hinton Canada 454k citations
  3. 3 Ross Girshick Israel 356k citations
  4. 4 Jun Wang China 340k citations
  5. 5 Xiangyu Zhang 327k citations
  6. 6 Xiaogang Wang China 260k citations
  7. 7 Yann LeCun United States 256k citations
  8. 8 Andrew Zisserman United Kingdom 255k citations
  9. 9 Li Fei-Fei United States 225k citations
  10. 10 Ilya Sutskever Canada 219k citations

Ranked by citations received across their whole record, among researchers with at least three works on this topic.

What are the most cited papers in Speech Recognition and Synthesis?

The most cited paper in Speech Recognition and Synthesis is "AI-Assisted Pipeline for Dynamic Generation of Trustworthy Health Supplement Content at Scale" (Kefallinos, Dionysios, Alexandris, Georgios, Maras, Alexis et al., 2018), with 46,030 citations. Next come "Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation" (25,290) and "A tutorial on hidden Markov models and selected applications in speech recognition" (22,958).

Most cited papers
#PaperReleasedCitations
1 AI-Assisted Pipeline for Dynamic Generation of Trustworthy Health Supplement Content at Scale Kefallinos, Dionysios, Alexandris, Georgios, Maras, Alexis et al. · arXiv (Cornell University) PreprintOpen access 11 Oct 2018 46,030
2 Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation Kyunghyun Cho, Bart van Merriënboer, Çağlar Gülçehre et al. Conference paperOpen access 2014 25,290
3 A tutorial on hidden Markov models and selected applications in speech recognition L. R. Rabiner · Proceedings of the IEEE 1989 22,958
4 Efficient Estimation of Word Representations in Vector Space Tomáš Mikolov, Kai Chen, Greg S. Corrado et al. · arXiv (Cornell University) PreprintOpen access 16 Jan 2013 18,196
5 Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data John Lafferty, Andrew McCallum, Fernando C. N. Pereira · ScholarlyCommons (University of Pennsylvania) Open access 28 Jun 2001 12,952
All 98 most cited papers, with filters and sorting

Where is Speech Recognition and Synthesis research done?

The largest centres of Speech Recognition and Synthesis research in 2022-now are Beijing (China), Tokyo (Japan), Seoul (South Korea) and Shanghai (China). Among places with at least 20 works in it, it is an unusually large share of all research in Redmond.

Largest cities, 2022-now

  1. 1 BeijingChina 520 works
  2. 2 TokyoJapan 213 works
  3. 3 SeoulSouth Korea 173 works
  4. 4 ShanghaiChina 166 works
  5. 5 ShenzhenChina 136 works
  6. 6 ChennaiIndia 120 works
  7. 7 Xi'anChina 107 works
  8. 8 BengaluruIndia 100 works
  9. 9 TaipeiTaiwan 100 works
  10. 10 HefeiChina 99 works

Where it is the local speciality

  1. RedmondUS · 27.5 works32×
← less than its size predicts1×more →

Location quotient: how much more of its research is in Speech Recognition and Synthesis than the world average.

See Speech Recognition and Synthesis on the map

Where is the best place to study Speech Recognition and Synthesis?

Among universities, judged by research, Carnegie Mellon University, Brno University of Technology and Chinese University of Hong Kong score highest, combining excellence, specialisation, size, growth and international reach. Research strength is one signal when choosing where to study; it does not measure teaching.

0%10%20%mean 18.74%fractional works in this field (log) →share of work in the world top 10% →Brno University of T…Carnegie Mellon Univ…Shanghai Jiao Tong U…Chinese University o…
above the meannear itbelow it

One dot per university in the first ten of the table below, at the default weights. The upper left is the interesting corner: small places doing unusually strong work.

Excellenceshare of its work in the world top 10%, 2022-now and all time averaged
40%
Specialisationhow much of its work is in this field, counted up to 10× the world share
10%
Sizeits output in this field, on a log scale
30%
Growthoutput growth, counted from -20% to +20%
10%
Internationalshare of its work with a co-author abroad
10%
Each slider sets a weight; the share it gets of the score is beside it.
#UniversityScoreTop 10%SpecialisationWorksGrowth
1Carnegie Mellon UniversityUnited States 77.119.6%16.5×54 +33.5%
2Brno University of TechnologyCzechia 71.321.1%18.2×23 +137.3%
3Chinese University of Hong KongChina 70.916.9%6.3×36 +37.9%
4Shanghai Jiao Tong UniversityChina 70.617.5%2.8×63 +85.6%
5University of Science and Technology of ChinaChina 69.713.9%7.9×73 +110.2%
6Imperial College LondonUnited Kingdom 69.723.8%2.8×22 +82.6%
7Nanyang Technological UniversitySingapore 68.018.9%4.7×29 +123.3%
8Northwestern Polytechnical UniversityChina 67.617.8%5.0×45 +67.0%
9Korea Advanced Institute of Science and TechnologySouth Korea 65.320.3%9.1×33 +99.8%
10Chinese University of Hong Kong, ShenzhenChina 65.017.6%15.2×22 +700.1%

Universities only. The score blends excellence (40%), size (30%), specialisation (10%), growth (10%), international reach (10%), 2022-now; growth compares 2018-21 with 2022-25 and counts up to +20%, since a jump from a small base is mostly noise.

Every institution in Speech Recognition and Synthesis, country by country

Is Speech Recognition and Synthesis research growing?

Output in 2022-2025 was 58% higher than in 2018-2021, peaking in 2025. The fastest-growing topics are Speech Recognition and Synthesis.

still counting19801990200020102020
grewheldshrank

The same series as a ribbon: one cell per year, darker for more. The line above answers how much; this answers when.

Its share of all research

Growing output can still be a shrinking share: all of science grows. This is Speech Recognition and Synthesis as a percentage of every work published that year, 0.041% in 2024.

still counting0.050%0.10%Speech Recognition and Synthesis0.041% in 202519801990200020102020

Fractional works of Speech Recognition and Synthesis over all fractional works that year. How the balance of science has shifted.

How much of it reaches the world top 10%

Year by year.

still counting5%10%15%Speech Recognition and Synthesis9.8% in 2025200020102020

Of each year's works in it, the share that reached the world's 10% most cited for their field and year. A field is the world for its own topics, so there is no other line to hold it against.

Which topics inside it are moving

Growth and decline on one axis around a shared zero. Two lists side by side hide the thing that matters: whether the growth dwarfs the decline, or the other way round.

Questions people ask

Who leads speech recognition research?

By volume in 2022-now, Chinese Academy of Sciences publishes the most Speech Recognition and Synthesis research, followed by University of Science and Technology of China and Centre National de la Recherche Scientifique. By volume, China and India publish the most (2.4k and 1.3k works in 2022-now).

All 1,273 questions in the Question Catalogue

Discussion

Ask a question or share a note about Speech Recognition and Synthesis. Posts are plain text and can be reported.