Speaker-Adaptive Speech Recognition Based on Surface Electromyography

Wand, Michael; Schultz, Tanja

doi:10.1007/978-3-642-11721-3_21

Cited by 11 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In (Wand and Schultz, 2009b) we reported first EMG recognition results based on 26 recording sessions with 13 speakers of the audible part of the EMG-PIT pilot study subset. For each speaker, the audible part of the SPEC set was used for training, and the BASE set for testing.…”

Section: Baseline Emg-based Speech Recognition Systemmentioning

confidence: 99%

“…Note that and (Wand and Schultz, 2009b) only used a stacking width of 5 frames. On the EMG-PIT corpus, the stacking width of 15 frames gives significantly better results (Wand and Schultz, 2009a).…”

Section: Baseline Emg-based Speech Recognition Systemmentioning

confidence: 99%

“…In (Wand and Schultz, 2009b) we considered speaker-dependent and speakerindependent phoneme-based EMG recognizers. This means that we regard each frame of the EMG signal as the representation of the beginning, middle, or end state of a phoneme.…”

Section: Modeling Phonetic Featuresmentioning

confidence: 99%

“…From traditional acoustic speech recognition it is known that modeling context-dependent phonemes reduces word error rates by about 30%. We investigated the use of triphones for EMG in (Wand and Schultz, 2009b), which gave a relative improvement of 11.5% for a speaker-independent EMG recognizer trained on about 77 minutes of EMG data. The limited success was certainly partly due to the lack of training data, since context-dependent modeling requires a large amount of data for reliable model estimates.…”

Section: Bundling Of Context-dependent Phonetic Featuresmentioning

confidence: 99%

See 3 more Smart Citations

Modeling coarticulation in EMG-based continuous speech recognition

Schultz

Wand

2010

Speech Communication

Self Cite

138

102

View full text Add to dashboard Buy / Rent full text

This paper discusses the use of surface electromyography for automatic speech recognition. Electromyographic signals captured at the facial muscles record the activity of the human articulatory apparatus and thus allow to trace back a speech signal even if it is spoken silently. Since speech is captured before it gets airborne, the resulting signal is not masked by ambient noise. The resulting Silent Speech Interface has the potential to overcome major limitations of conventional speech-driven interfaces: it is not prone to any environmental noise, allows to silently transmit confidential information, and does not disturb bystanders.We describe our new approach of phonetic feature bundling for modeling coarticulation in EMG-based speech recognition and report results on the EMG-PIT corpus, a multiple speaker large vocabulary database of silent and audible EMG speech recordings, which we recently collected. Our results on speaker-dependent and speaker-independent setups show that modeling the interdependence of phonetic features reduces the word error rate of the baseline system by over 33% relative. Our final system achieves 10% word error rate for the best-recognized speaker on a 101-word vocabulary task, bringing EMG-based speech recognition within a useful range for the application of silent speech interfaces.

show abstract

Section: Baseline Emg-based Speech Recognition Systemmentioning

confidence: 99%

Section: Baseline Emg-based Speech Recognition Systemmentioning

confidence: 99%

Section: Modeling Phonetic Featuresmentioning

confidence: 99%

Section: Bundling Of Context-dependent Phonetic Featuresmentioning

confidence: 99%

See 2 more Smart Citations

Modeling coarticulation in EMG-based continuous speech recognition

Schultz

Wand

2010

Speech Communication

Self Cite

138

102

View full text Add to dashboard Buy / Rent full text

show abstract

“…Some proposed systems focus on recognizing words from isolated SEMG signals [2] [3] [4] [5] [6] [7]; these systems demonstrate the feasibility of using SEMG signals to improve the performance of conventional speech recognition systems in some noisy environments [2] [8]. The improvement is mainly due to the fact that SEMG signals are immune to acoustic noise.…”

Section: Introductionmentioning

confidence: 99%

Non-acoustic Communication with Speech Smoothing

Lam

2012

SIPIJ

View full text Add to dashboard Buy / Rent full text

Speaker-Adaptive Speech Recognition Based on Surface Electromyography

Cited by 11 publications

References 13 publications

Modeling coarticulation in EMG-based continuous speech recognition

Modeling coarticulation in EMG-based continuous speech recognition

Non-acoustic Communication with Speech Smoothing

Contact Info

Product

Resources

About