Method for the subjective assessment of intermedi-ate quality levels of coding systems

By Sugato Chakravarty. 2001

361 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

WORLD: A Vocoder-Based High-Quality Speech Synthesis System for Real-Time Applications (2016, methodology): “The subjective evaluation was based on MUSHRA (MUltiple Stimuli with Hidden Reference and Anchor) [32], and it and the processing speed evaluation compared WORLD against Legacy-STRAIGHT and TANDEM-STRAIGHT.”
Real-Time Speech Frequency Bandwidth Extension (2020, methodology): “To assess perceptual quality, we perform subjective evaluation using the MUSHRA methodology [25] with 10 VCTK test audios (2–5 seconds each).”
High-quality Speech Coding with Sample RNN (2018, methodology): “The subjective quality is evaluated with a methodology based on MUSHRA[12].”
A Deep Generative Architecture for Postfiltering in Statistical Parametric Speech Synthesis (2015, methodology): “We evaluated the methods in Table I using the MUltiple Stimuli with Hidden Reference and Anchor (MUSHRA) methodology [40].”
Neural Music Synthesis for Flexible Timbre Control (2018, methodology): “MOS allows a simple interface that is more suitable for non-expert listeners in a crowdsourcing setup than e.g. MUSHRA [26] and is used as a standard approach for assessing the perceptual quality of WaveNet syntheses [16, 27].”
PLATINUM: A method to extract excitation signals for voice synthesis system (2012, methodology): “To verify the eﬀectiveness of the proposed system, a MUSHRA-based evaluation [10] was carried out.”
Optimally using the Bluetooth subband codec (2010, background): “1534 [13] describes a procedure on how to judge the impact of intermediate audio degradations.”
User measures of quality of experience: why being objective and quantitative is important (2010, methodology): “Even if the five scale positions of Bad, Poor, Fair, Good, and Excellent are numbered from 1 to 5 [2] or from 0 to 100 [3], the numbers may not accurately reflect the distance between the scale positions, so it may be invalid to calculate statistics such as the mean, standard deviation, and correlat”
Efficient coding of natural sounds (2002, methodology): “We further used a reduced number of audio coding schemes (MPEG-2/4 AAC, CSTQ-AAC and DSTQ-AAC) to allow the evaluation of the perceived audio quality with the resource-demanding subjective listening test method MUSHRA (multi stimuli with hidden reference and anchor points) [67].”
A Streamwise Gan Vocoder for Wideband Speech Coding at Very Low Bit Rate (2021, unknown): “We conducted a subjective listening test following the ITU-R MUSHRA [23] recommendation comparing classical and neural speech coders.”