On June 7th I met with Soundcloud’s Ben Fawkes. Ben is the Audio Content Manager and handles business development for everything that is non-music related (e.g. spoken word, poetry, podcasts, comedy, et cetera). The purpose of the meeting was to discuss the Spoken Web project in detail, with a focus on the features we are trying to develop, in order to see if and how the Soundcloud platform could be used to to create these features.
Several members of the SpokenWeb team raised specific questions for me to discuss with SC. Before I get to those specific questions, I would sum up our meeting like this:
- SoundCloud is very interested in and enthusiastic about our current project (and also future related and expanded projects).
- Many of the features we are requesting are fairly specific, niche features that aren’t currently in demand by the majority of their users.
- SC has limited resources for development. In most cases, they probably can’t develop the features we want (i.e. do the work) but are willing to support us doing the development using their API in any way that they can (i.e. providing support with their API and technical support teams, pointing to similar sites, projects, open code, etc). Ben mentioned that he is personally interested in keeping in touch about the project and would provide help and guidance wherever he could.
- SoundCloud is also interested in helping us promote our site(s) and digital archives, should they become open to the public e.g. via Facebook, the SoundCloud blog, et cetera.
- SC is currently putting a lot of energy into podcasts (and making them easier to distribute, “freeing them” from Apple’s closed distribution) and also developing SC into something like the audio equivalent of Twitter (i.e. making it easier for people to instantly record and publish short bursts of audio on the web). Some examples, question and answer audio sites (also see http://takesquestions.com/), syncing your voicemail and SC accounts to enable easy publishing of audio on the web (see also http://importer.soundcloudlabs.com/), et cetera.You can also watch SC CEO Alexander Ljung’s presentation at The Next Web for more details. Finally, there are an increasing numbers of apps on the SC site that make it easy to do mobile recordings (e.g., from smartphones) that are automatically uploaded to your SC account.
- Ben also suggested that we consider applying to attend a Music Hack Day to try and solve some of our design problems. Basically you apply, describe a problem, and then if you’re accepted groups of people work on developing solutions over a couple of days.
And now to some of the specific features and issues that we discussed.
- Two-way tethering of audio playback and transcripts. A great example of 2-way tethering using SC is demonstrated by this site: http://hyper-audio.org/r/. The development was done by Mozilla so the code should be open and available if we contact them. I also showed Ben how we were currently working on setting up the tethering in our WordPress site. He mentioned that we should feel free to contact “firstname.lastname@example.org” anytime we need help, and that one piece of advice that they offered so far was: “you should press play and then pause on the player and allow it to buffer”. It doesn’t mean anything to me right now but I’m listing it anyways. Ben also mentioned that SC is in advanced discussions with http://speakertext.com/ to try and integrate some form of automated transcriptions into the standard interface. Given all of this, I’m thinking we should consider moving away from using the “comment bubbles” for sections of the transcript (since there are numerous problems, and that we would need to solve them on our own more or less) and instead try moving towards something like the http://hyper-audio.org/r/ solution.
- Different types of sound visualization (e.g., colour coding the waveform to indicate different sections, poems, voices, etc; other visualizations, such as pitch representation, in addition to the waveform). Basically SC doesn’t have any plans to develop a feature like this for the standard interface but would try to support us developing this type of feature using the API.Ben also mentioned that we could consider using “Sets” within SC which allows you to break up a longer recording containing multiple poems (for example) into its component parts.
- Platform for network of spoken archives. I mentioned the possibility of a future network of spoken word archives, and SC said that they were interested in being involved since their platform was designed for exactly this type of thing.
- Geolocation-related features. I showed Ben Celyn’s recent work with Google’s My Maps tool. He mentioned that SoundCloud recently collaborated with the SouthBySouthWest music festival to create something similar. Their tool stamps the location onto a recording done at a particular venue at the SXSW festival and then allows playback at that specific spot on the map. http://soundsbysouthwest.com/. SC also have a number of projects involving integration with FourSquare.
- Adding timed comments by numerical value instead of clicking on timeline. I asked Ben Darren’s question about this. Ben mentioned that other clients have had problems with this but they do not currently have any plans to change this.
- Manipulation of audio playback. For example, slowing audio without pitch change, “audio quotation” (i.e. designation of a section of the audio to be looped for some period of time), presentation of multiple waveforms at same time). Ben mentioned for the most part there were no plans to build these kinds of features into the general interface. He did mention that one of his friends is working on integrating a basic web-based digital audio editor with SC and that he will share it with us when it is done.
- Organizing large numbers of recordings in SC.When Ben and I were discussing the possibility of expanded projects that use large numbers of recordings, he mentioned that we should be aware of SC’s desktop client which is more powerful for organizing files than the SC web site.