- From: Olivier Thereaux <olivier.thereaux@bbc.co.uk>
- Date: Mon, 19 Mar 2012 13:35:21 +0000
- To: Joseph Berkovitz <joe@noteflight.com>, Chris Lowis <chris.lowis@bbc.co.uk>, Thierry MICHEL <tmichel@w3.org>
- CC: public-audio@w3.org
- Message-ID: <4F673619.50400@bbc.co.uk>
Hi Joe, all. On 12/03/2012 15:23, Joseph Berkovitz wrote: > The important thing about the metronome UC is with respect to recording > and audio acquisition, as Olivier points out, not sound generation. It > becomes important to be able to correlate the timing of audio input with > respect to the timing of audio output at a sample-accurate level. I did > not look at the UC yet but if it can include audio recording/analysis of > someone playing along with the generated metronome, then this will make > it a distinct and useful UC. Based on the discussion so far, here is a proposed new wording for UC-13. UC-13: Web-based guitar practice service A serious guitar player uses a web-based tool to practice a new tune. Connecting a USB microphone and a pair of headphones to their computer, the guitarist is able to tune an acoustic guitar using a graphical interface, set a metronome to keep the tempo then start recording a practice session. The audio input from the microphone is automatically analysed to detect whether the musician is keeping a regular beat. The music played during the session is recorded and can be saved to a variety of file formats locally or on the online service where others can replay, comment on the performance and annotate any section to help the musician improve technique and delivery. The original text for UC-13 is here: http://www.w3.org/2011/audio/wiki/index.php?title=Use_Cases_and_Requirements&oldid=422#UC-13:_Recording_and_Saving_Audio_Recordings I gather that this new text would add the following requirements: * Sample-accurate scheduling of playback (for the metronome) * Level detection (would be used as part of tempo detection?) * Frequency domain analysis (for tuner) Joe mentioned that “It becomes important to be able to correlate the timing of audio input with respect to the timing of audio output at a sample-accurate level” - should that become a separate requirement? Thoughts? I think it is a richer use case than the simple "recording and replaying voice" but there may be a key aspect of the "old" UC-13 I am missing. Thanks, Olivier
Attachments
- application/pkcs7-signature attachment: S/MIME Cryptographic Signature
Received on Monday, 19 March 2012 13:35:52 UTC