Rec. ITU-T P.501 (04/2025) - prepublished version - Table of Contents
1 Scope
2 References
3 Definitions
     3.1 Terms defined elsewhere
     3.2 Terms defined in this Recommendation
4 Abbreviations and acronyms
5 Conventions
6 Overview of test signals and typical applications
7 Types of test signals
     7.1 Non-speech-like (fully artificial) signals
          7.1.1 Deterministic signals
               7.1.1.1 Description
               7.1.1.2 Application
          7.1.2 Random signals
               7.1.2.1 Description
               7.1.2.2 Application
          7.1.3 Combined random and deterministic signals
               7.1.3.1 Description
               7.1.3.2 Application
     7.2 Speech-like signals
          7.2.1 Composite source signals (composed signals in time)
               7.2.1.1 Description
               7.2.1.2 Practical realization of a composite source signal for measurements up to 20 kHz (fullband)
               7.2.1.3 Application
               7.2.1.4 Fullband composite source signal for double-talk
               7.2.1.5 Band limitation of fullband composite source signals and speech-like power density spectrum
               7.2.1.6 Narrow-band composite source signal with speech-like power density spectrum
                    7.2.1.6.1 Description
                    7.2.1.6.2 Application
          7.2.2 Speech-like modulated noise
               7.2.2.1 Description
               7.2.2.2 Application
          7.2.3 Composed signals in frequency (probe tone technology)
               7.2.3.1 Description
               7.2.3.2 Application
          7.2.4 Voice-like composed signals in frequency
               7.2.4.1 Description
               7.2.4.2 Application
          7.2.5 Complex composed signals
               7.2.5.1 Simulated speech generator
                    7.2.5.1.1 Description
                    7.2.5.1.2 Application
               7.2.5.2 Artificial voice [ITU-T P.50]
               7.2.5.3 Artificial conversational speech [ITU-T P.59]
               7.2.5.4 Speech-model process controlled by discrete Markov chains
                    7.2.5.4.1 Description
                    7.2.5.4.2 Application
     7.3 Speech signals
          7.3.1 Reference speech samples
          7.3.2 Speech signals for single-talk testing
               7.3.2.1 Description
               7.3.2.2 Application
          7.3.3 Compressed speech signals for testing
               7.3.3.1 Description
               7.3.3.2 Application
          7.3.4 Short words for activation (temporal) tests
               7.3.4.1 Description
               7.3.4.2 Application
          7.3.5 Speech signals for double-talk testing
               7.3.5.1 Description
               7.3.5.2 Application
          7.3.6 Speech sequences for echo performance testing
               7.3.6.1 Description
               7.3.6.2 Application
          7.3.7 Conditioning speech sequences
               7.3.7.1 Description
               7.3.7.2 Application
          7.3.8 Filters for limiting the speech test signal bandwidth
     7.4 Additional languages
          7.4.1 Chinese speech samples
               7.4.1.1 Chinese reference speech samples
               7.4.1.2 Chinese single-talk speech sequence
               7.4.1.3 Chinese double-talk speech sequence
               7.4.1.4 Chinese conditioning speech sequence
Annex A  Test signals for terminal coupling loss tests
Annex B  Speech files and noise sequences
     B.1 General
     B.2 Description of the recording procedure used for speech signals
     B.3 Test sentences
          B.3.1 Chinese (fullband)
          B.3.2 Dutch (fullband)
          B.3.3 English (fullband)
          B.3.4 English (American)
          B.3.5 Finnish (fullband)
          B.3.6 French (fullband)
          B.3.7 German
          B.3.8 German (fullband)
          B.3.9 Italian (fullband)
          B.3.10 Japanese (fullband)
          B.3.11 Polish
          B.3.12 Spanish (American)
     B.4 Noise sequences
          B.4.1 Binaural noise recordings
          B.4.2 Monaural noise recordings
Annex C  Speech files prepared for use with ITU-T P.800 conformant applications and perceptual-based objective speech quality prediction
     C.1 General
     C.2 Test sentences
          C.2.1 Dutch (fullband)
          C.2.2 Chinese (fullband)
          C.2.3 British English
          C.2.4 Finnish
          C.2.5 French
          C.2.6 German
          C.2.7 Italian
          C.2.8 Japanese
Annex D  Speech files composed of a pair of sentences spoken by a male and a female speaker
     D.1 General
     D.2 Test sentences
          D.2.1 Dutch (DU)
          D.2.2 British English (EN)
          D.2.3 Finnish (FI)
          D.2.4 French (FR)
          D.2.5 German (GE)
          D.2.6 Italian (IT)
          D.2.7 Chinese (Mandarin) (CN)
          D.2.8 American English (AM)
          D.2.9 Japanese (JP)
Annex E  Representative short speech sample for technical measurements
     E.1 General
     E.2 Description
Appendix I  Description of the processing applied to the speech signals in clause 7.3
     I.1 Filter for DC removal
     I.2 Creation of the single-talk speech sequence
     I.3 Example high-pass filter designs
Appendix II  ITU-T P.863 results on ITU-T P.501 and ITU-T P.565 speech samples
     II.1 Reference speech samples in ITU-T P.501 Annex C and Annex D
     II.2 Processing and scoring of speech samples for offline reference conditions
     II.3 ITU-T P.863 scores on Annex C and Annex D samples on offline reference conditions
     II.4 ITU-T P.863.1 average scores on Annex C samples
     II.5 Results obtained with the speech sample from [ITU-T P.565.1]
     II.6 Summary
Bibliography