One of the hard tasks that lies ahead for the Hypervoice Consortium is to define what the ‘Hypervoice’ concept actually is. If wedraw it too wide, it will turn into a marketing term: fully press-release compliant with “social enterprise 2.0”, but of no substance whatsoever. Too narrow, and we will exclude too many use cases and community members.
Our initial working definition has five defining characteristics required for a technology or product to be considered fully within the Hypervoice framework.
1) Activity stream integration
The central essence of Hypervoice is ‘linking what we say to what we do’. Since ‘what we do’ is synonymous with the creation of an activity stream, ‘what we say’ has to be part of that stream too. So while a phone may have a call log, recording ‘who we talked to’, it is divorced from our wider digital activity stream. In this example, the voice content and its metadata are still silo-based data. Hypervoice requires ‘what we say’ to be part of a larger, integrated hypermessaging framework.
The second defining characteristic of the Hypervoice model is that the spoken audio is recordable. It should always be possible to access the recorded spoken context where it is available and access rights allow it. However, that isn’t the same as it always having been actually or completely recorded. For example, those present on a conference call may choose to not record the joking about the office party before the main meeting begins. Indeed, even if a call is not recorded at all, it is enough to have captured the mere existence of the call being initiated from within a larger hypermessaging system. This enables the final product: notes, slides and customer records tied together into a searchable whole.
Voice is a naturally social medium, and one through which we share our thoughts and feelings. A Hypervoice conversation is about using the amplification power of computers to share those conversations more widely over time and space. That not only means that snippets of recorded audio can be shared, but also that whole Hypervoice conversations can be shared. We should be able to syndicate any subset of the data for others to subscribe to (e.g. conversations about a particular customer account or trouble ticket).
4) Workflow integration
Hypervoice is a purposeful medium: it is about getting work done by turning intent into action. A Hypervoice conversation ought to have an inbound intent (why are we talking?) and an outbound action (what next?). The inbound intent could be as simple as a subject line, or could be an agenda, or even a rich media object like a project plan being reviewed. The outbound action could also take many forms: tagged audio, unstructured to-do items, or structured decisions such as loan approvals or declines. The key is that the outcome must improve productivity in a meaningful and significant way by, for example, eliminating rework or redundant meetings. A Hypervoice-enhanced meeting must be superior to a face-to-face meeting in a tangible, quantitative way.
Not all conversations are searchable today in the original voice, since transcriptions are not universally available or perfect. Furthermore, work is done in many languages, and these may not be readily machine-translatable. Hypervoice therefore encompasses other technologies that allow for the mark-up of recorded audio. Social gestures, such as tagging and liking, attach active meaning to the recorded audio. Passive tagging can happen via slide-sharing, note-taking and other natural, automatic meeting behaviors.
Hypervoice is language-agnostic, and always brings you back to the original voice content (where available), and not simply a text transcription. The key is to have both media and metadata available. If only the recorded media is available, it would be voice, not Hypervoice.
Community Feedback Welcome
At this early stage, we are very aware of the potential for shifting these boundaries, and changing what is ‘necessary’ and ‘sufficient’ for the definition of the Hypervoice model. This is most certainly a “first draft” and I would be very glad to receive your feedback in the Comments section.