Thank you Tom, sadly, you are correct. Performers often take obscene liberties with clearly specified time signatures or stylistic practices, resulting in totally bizarre tempos of performances and recordings.
Rythmic signatures are a different matter all together, and are subject to their own special kind of manglings.
Eldartford, you are right of course, about time domain compression/expansion. I have played with analog time domain compression of speech streams since the mid 70s. . . . and in those days results were pretty horrendous. Now days, using advanced algorithms, reasonable intellegibility can be achieved by speeding up digital recordings of speech up to 3X. Text to speech engines (TTS) can yield intelligible speech upwards of 500WPM by adopting differing compression scales for different phonemes. I have heard formant-based TTS engine actually running at 1200WPM, but do not pretend to understand them at such speech rates. G.
Rythmic signatures are a different matter all together, and are subject to their own special kind of manglings.
Eldartford, you are right of course, about time domain compression/expansion. I have played with analog time domain compression of speech streams since the mid 70s. . . . and in those days results were pretty horrendous. Now days, using advanced algorithms, reasonable intellegibility can be achieved by speeding up digital recordings of speech up to 3X. Text to speech engines (TTS) can yield intelligible speech upwards of 500WPM by adopting differing compression scales for different phonemes. I have heard formant-based TTS engine actually running at 1200WPM, but do not pretend to understand them at such speech rates. G.