METHODS FOR SUBJECTIVE DETERMINATION OF TRANSMISSION QUALITY Summary

Citations in the corpus

• Low Bit-rate Speech Coding with VQ-VAE and a WaveNet Decoder (2019, background): “MOS evaluations [21] map ratings from bad to excellent into the range 1-5 with 0.”
• Composition of Deep and Spiking Neural Networks for Very Low Bit Rate Speech Coding (2016, methodology): “The test consisted of 20 randomly selected utterances from the Nancy test data, at least 2 seconds long.”
• GELP: GAN-Excited Liner Prediction for Speech Synthesis from Mel-spectrogram (2019, methodology): “Figure 2 shows the test results as mean opinion scores (MOS) [34] with the 95% confidence intervals, along with stacked score distribution histograms.”
• Perceptual Objective Listening Quality Assessment (POLQA), The Third Generation ITU-T Standard for End-to-End Speech Quality Measurement Part I-Temporal Alignment (2013, methodology): “These data consist of reference and degraded speech files with a subjective quality score on each degradation [1], [2].”
• GlotNet—A Raw Waveform Model for the Glottal Excitation in Statistical Parametric Speech Synthesis (2019, methodology): “A category comparison rating (CCR) [56] test was performed to evaluate the synthetic speech quality.”
• Speaker-independent raw waveform model for glottal excitation (2018, methodology): “To evaluate the subjective quality of the synthetic speech, we conducted pairwise category comparison rating (CCR) tests [26], where the listeners were presented with a pair of samples and asked to rate the comparative quality on a 7-level scale, ranging from -3 (much worse) to 3 (much better).”
• Handover decision using fuzzy MADM in heterogeneous networks (2004, methodology): “The rating score can be further converted to MOS.”
• Perceptual Objective Listening Quality Assessment ( POLQA ) , The Third Generation ITU-T Standard for End-to-End Speech Quality Measurement Part II – Perceptual Model (2013, methodology): “Within the telecom industry one mostly uses a five-point opinion scale [1], [2] (see Table 1).”
• GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-spectrogram (2019, methodology): “The MOS score measured for the WaveNet vocoder in the current study may seem low compared to the score (about 4.5) reported in [8].”
• A Non-Intrusive Quality and Intelligibility Measure of Reverberant and Dereverberated Speech (2010, background): “Commonly, subjective quality tests have listeners rate the quality of the speech signal on a pre-specified scale [5].”