Using Audacity to make podcasts

[KenZakreski]

I note AAC is the format Apple uses to provide chapters, which might allow the display of songs titles and artist names , as they are being played, on regular players, which might work to allow song logging for a computerized logging system.

That might work on some iPods, but there is no guarantee that it will work on other brands, or even on all iPods. This is an inherent problem of non-standard formats.

This seems to be the feature:
http://coverville.com/archives/interest ... eband-mac/

This is definitely a feature for AAC files only.

Although that article implies this feature is only for iPod/iPhone/iPad, I suspect chapters may be visible in iTunes, or at least you can navigate to them. Can you see GarageBand chapters on iTunes in OS X, Ken?

Gale

Yes this is a feature for AAC files.

It is reported that GarageBand chapters do work in iTunes and iTunes will track songs as played in Itunes, which we are currently using. I cannot confirm this as I am away from our studio.

Due to the 12,000 song limit (time bomb) in iTunes we are promptly vetting AirTime as our station scheduler and player for pre recorded radio shows.

We have applied for funding to modify AirTime to use under the Canadian logging requirements which include CanCon, hits, etc. this way we can log the songs played.

Our program director wants to permit listeners to skip through podcasts so they may listen to their favourite shows from a daily download. This feature may be a by product of chapters.

Ken
President
CKGI.ca 98.7fm

[KenZakreski]

Like most other Canadian Community Radio Stations we support audacity for building radio shows.

Thanks for your support, Ken.

When I build my radio show I drag the audio tracks into Audacity, the meta data seems to come with it. The name of the song goes to the track title in Audacity.

I would like the meta data from each song to stay in the final audio file for all the songs included. This would serve two purposes, when my show is podcasted the listener would see the song names and not just the show names. Also, we could log the songs played in a automation service so I wouldn't have to type in the songs played again.

Not sure if this is called 'chapters' like Garage Band uses or Label Track in the Audacity manual.

I am not sure what you mean by "meta data". Have you enabled "Show Metadata Editor prior to export step" in Import / Export Preferences ? Here is the page about Metadata Editor: http://manual.audacityteam.org/o/man/me ... ditor.html .

Yes I have enabled "Show Meta Data... " and noticed the last imported track's information is stored int he metadata container.

By metadata I am referring to the audio tracks ID3v2 tags stored with the file. Additionally, in Canadian broadcasting we are
Required to log for Canadian Content, some stations also include fem con (an equity term describing female to male ratio involved with the recording), if the song is a hit as describe in the regulations, etc. the additional information is input by trained volunteers for those stations with digital libraries.

If you are importing multiple songs into separate Audacity tracks, then only the metadata (in the sense above) for the last imported song is retained. In any case, if you make all those songs into one Audacity track and export it as one audio file, you can't make separate sections within the one file with metadata for each song.

Yes, this is my understanding. We would appreciate being able to retain the metadata for each track to remain and be displayed at the instant the song is played. Using them as chapters would be a secondary feature we would appreciate.

Have you tried Splitting a recording into separate tracks - labelling each song, then using File > Export Multiple to export a file for each song? Then you can give each file its own metadata and you can make a playlist comprising the file for each song. This will work fine if you are streaming, but not if you want to give the listeners a single file to download from a web site.

We start with separate track, the songs themselves are distinct. Where things get complex is some programmers with advanced skills will often speak while the track is playing. We can omit logging the spoken word portion unless they mention the radio station, station IDs (StnIDs) which we must do regularly and must log. We can code the meta data for StIDs and they can also be logged.

Again, if you mean metadata like genre and artist name, you will have to type it for each song, because Audacity only retains the metadata for the last file.

Yes, I mean metadata like genre , artist name, title, CanCon and other required information.

[Gale Andrews]

Our program director wants to permit listeners to skip through podcasts so they may listen to their favourite shows from a daily download. This feature may be a by product of chapters.

Yes players that support Apple chapters can definitely skip between them.

Players that support ISO chapters will probably be able to skip but I don't think the Apple and ISO chapters are interchangeable features.

If you are offering the Audacity developers commercial rates for implementing AAC bookmarks/chapters and/or a way to capture metadata from multiple imported files, then you should write to our feedback address with a clear description of the features you wish to see. For example, is Track Title and Track Number all you want to pass to each "chapter"?


Gale

[KenZakreski]

I would like the meta data from each song to stay in the final audio file for all the songs included. This would serve two purposes, when my show is podcasted the listener would see the song names and not just the show names. Also, we could log the songs played in a automation service so I wouldn't have to type in the songs played again.

There is one way of doing something like that which is widely supported by many players, and that is to use a playlist for the show. (http://en.wikipedia.org/wiki/Playlist#O ... e_Internet)

The way it works is that each track is a separate file, (with its own metadata). The files are organised into a sequence by a special text file (such as an "M3U" file: http://en.wikipedia.org/wiki/.m3u). The audience download the playlist and their media player then drags in each of the songs, jingles, introductions, etc. in sequence.

There are several benefits to this method:
It is widely supported by many media players.
It is extremely easy to change the program at short notice if necessary (just edit the playlist file).
Most players are able to start playing almost instantly after the (very small) playlist file has been downloaded.
Song information can be written directly into the playlist, so the song titles will usually appear even if the media player does not support the metadata format.
No need to handle huge files - each track is just normal song size.
Easy to make an audio CD of the show because each "track" is a separate file.
Playlist creation can be integrated into a database application.
Supports multiple file formats so can reduce the time spent transcoding files.

Yeah, we use playlists, iTunes. What would be nice if there was an easy way to make the podcast for internet distribution. What I can do is run the playlist through NiceCast which records in real time. Audacity makes the podcast as fast as it can, usually minutes (and we all thank you for that)

Can Audacity rip an mp3 podcast from a playlist?

And not sure about "media player then drags in each of the songs, jingles, introductions, etc. in sequence." Our digital library will store music as high quality audio files for broadcast (possibly lossless), how will they be converted to internet grade mp3?
And how does the listener get access to the files, which reside on a NAS box with very nervous on air programmers wanting responsive access to a digital library , could be a problem with public access to our digital library.

Ken
president
CKGI.ca 98.7fm

[KenZakreski]

Merged the duplicate topics and deleted the duplicate post. Please only post once.

Sorry, I saw this topics had low traffic and thought my question should be posted in the other topic.

If this is a feature request, what is it for?

This feature will allow display and logging of songs within a single audio file such as a podcast. Enough metadata tags are included to permit Canadian Radio logging requirements

A second consequence of this feature is the use of chapters will permit spot to spot scrubbing searching within the audio file.

That projects should store metadata per track?

The song information (metadata) is retained when imported to Audacity and in the exported file.

In your case, when you want to make a single exported file, is the request also that the label should store the metadata - that you should be able to link the metadata that came with the file to a label?

A label may not allow the structure of the metadata tags. Are labels like a vorbis containers?

Both those were proposed on audacity-devel a few years ago, but there is no patch for it.

What did that proposal look like?
Maybe now that Apple has chapters there is certainty in the market what chapters should look like.

Or is getting the name of the imported file into the label name all that concerns you?

If the label is retained during the export process and can be read by a machine to determine the song played , then yes. My understanding that that is not the case.

Ken

[KenZakreski]

Our program director wants to permit listeners to skip through podcasts so they may listen to their favourite shows from a daily download. This feature may be a by product of chapters.

Yes players that support Apple chapters can definitely skip between them.

Players that support ISO chapters will probably be able to skip but I don't think the Apple and ISO chapters are interchangeable features.

If you are offering the Audacity developers commercial rates for implementing AAC bookmarks/chapters and/or a way to capture metadata from multiple imported files, then you should write to our feedback address with a clear description of the features you wish to see. For example, is Track Title and Track Number all you want to pass to each "chapter"?


Gale

Right now I am surveying other community and campus stations to find out if there is agreement on metadata tag structure. Fem con (denoting a female component in the songs creation) is not a Regulatory requirement but important none the less. Some stations support that. Other tags may be coming into usage we are not yet aware of. This is a brave new world.

File format is not yet agreed on widely across our sector for the digital libraries. Some stations have agreed to high quality mp3 , other use ogg Vorbis, I like AIFF, I mean why compress?

There is discussion that a unique song identifier is all we need, this is a single tag in the metadata structure.

The Canadian National Campus Community Radio Conference (NCRC) is set for June 2nd to the 7th Victoria BC. Maybe you can send someone and help us sort all this out. The de facto standard for ID3v2 might be all we need, but then. I'll be presenting on marketing and can intro you to Johnnie.

There is usually a lot of beer and cool bands if you like that kind a thang.

In Canada we are funded by the Community Radio Fund of Canada (CRFC) and software development is now an approved expense. We applied to get AirTime vetted for use. Maybe some monies available.

What is the community radio rate for audacity development? Do developers like beer and loud music?



Ken
CKGI.ca 98.7fm

[steve]

And not sure about "media player then drags in each of the songs, jingles, introductions, etc. in sequence." Our digital library will store music as high quality audio files for broadcast (possibly lossless), how will they be converted to internet grade mp3?
And how does the listener get access to the files, which reside on a NAS box with very nervous on air programmers wanting responsive access to a digital library , could be a problem with public access to our digital library.

It depends on your infrastructure. These are really questions for your technical guys.

Uploading high quality lossless files to the file server could pipe the files through an encoder, creating a "broadcast ready" compressed copy on the web server.
As users will mostly only downloading those parts of the show that they listen to, it could vastly reduce the amount of (costly) bandwidth at the expense of slightly increasing the (cheap) number of server requests. For security, your uncompressed originals should of course be isolated from the public network.

[KenZakreski]

And not sure about "media player then drags in each of the songs, jingles, introductions, etc. in sequence." Our digital library will store music as high quality audio files for broadcast (possibly lossless), how will they be converted to internet grade mp3?
And how does the listener get access to the files, which reside on a NAS box with very nervous on air programmers wanting responsive access to a digital library , could be a problem with public access to our digital library.

It depends on your infrastructure. These are really questions for your technical guys.

Uploading high quality lossless files to the file server could pipe the files through an encoder, creating a "broadcast ready" compressed copy on the web server.
As users will mostly only downloading those parts of the show that they listen to, it could vastly reduce the amount of (costly) bandwidth at the expense of slightly increasing the (cheap) number of server requests. For security, your uncompressed originals should of course be isolated from the public network.

The lower quality public distributions of our radio shows (podcasts) are stored on third party hosts ( CKGI uses the NCRA Exchange site and SoundCloud ). These are ripped to mp3 as needed, not at the time they are entered into our digital library.

[steve]

Even if AAC metadata works for your listener's media players, are they supported by NCRA Exchange and SoundCloud?

I'm not trying to tell you how to run your show, I'm just putting forward some possible solutions that you may not have previously considered.

I had a quick look at the NCRA Exchange.
I clicked on a "Listen" link and it downloaded a M3U file.

This is the contents of the file:

Code: Select all

#EXTM3U
#EXTINF:50, Endeavours - Commitment to Change
http://previous.ncra.ca/business/admin_ncra/progex/programFiles/1321/endeavours_-_feb_17.mp3

So it seems that they already use playlists.

[KenZakreski]

Even if AAC metadata works for your listener's media players, are they supported by NCRA Exchange and SoundCloud?

I'm not trying to tell you how to run your show, I'm just putting forward some possible solutions that you may not have previously considered.

I had a quick look at the NCRA Exchange.
I clicked on a "Listen" link and it downloaded a M3U file.

This is the contents of the file:

Code: Select all

#EXTM3U
#EXTINF:50, Endeavours - Commitment to Change
http://previous.ncra.ca/business/admin_ncra/progex/programFiles/1321/endeavours_-_feb_17.mp3

So it seems that they already use playlists.

Yes, but the NCRA Exchange is usually a single long file mp3... Not sure whats up.

Think about the programmers, who after making the show need to deal with what we decide. And have to promote the show as well and plan the next episode.


The real thing is, can the station log a song from a pre recorded show? If I build it as a playlist the the delay is making a podcast. If I make a podcast and the songs can't be logged I need to build a playlist. It would be desirable to build a precorded show, export a file for the station to play ( and get logged) then export an mp3 for podcast distribution.

What is easier?

Ken