allmusic.com - one of the largest repositories of music reviews in the world. If you want to make a massive corpus, I would just scrape that.
pitchfork.com - has (indie) music reviews going back to 1999, and they have 5 or so album reviews a day. The album reviews section is here: http://pitchfork.com/reviews/albums/ They changed their site format a few months ago, but before that I scraped all the reviews to make a music review corpus. I could zip that up and send it to you, though it may require some post-processing here and there. There are over 10,000 reviews in that scrape.
metacritic.com - is a review aggregator site for movies, music, games, etc. It has links to reviews on other websites and then normalizes the scores from each website to give a composite score.
nme.com / spin.com / rollingstone - all have music reviews on their website, another good source for webscraping.
Joel
------------------------------
Message: 5 Date: Thu, 10 Nov 2011 14:51:38 -0500 From: Keely <km.mimnagh at gmail.com> Subject: [Corpora-List] North america newspaper corpus To: corpora at uib.no
Hi I am a master's student. I am running a study on the language of music critics. Does anyone know of a corpus that breaks down newspapers by sections. So I can parse on music reviews?
Any help would be much appreciated.
Keely
-- Keely Mimnagh M.A. Candidate Music and Culture Carleton University Ottawa, Ontario