Method and apparatus for speech behavior visualization and gamification
The disclosed system and methods provide real-time information about how agents and customers sound as they are speaking, allowing a supervisor to continuously monitor how agents are doing. The system allows agents to visualize their own speech behavior performance during and after a conversation while viewing important comparative information about prior conversations, and gamifies conversations in real-time by providing visual comparison between the live conversations and various target metrics. The visualization in-turn enhances agent interactive skills such as active listening and mirroring, as well as decision-making skills based on observations of customer engagement and distress levels.
1. A gamification system for presenting analysis of speech audio signals in real-time, the system comprising:
an application server configured to provide computed streamed behavioral signals to a computing device of an agent,
wherein the application server is connected to a real-time signal processing system configured to apply metrics to transform streamed audio conversational signals into the computed streamed behavioral signals at low-latency;
wherein the real-time signal processing system is connected to a real-time audio capture system configured to receive source conversational audio signals of current audio involving the agent and convert the received audio signals into analyzable digital audio as the streamed audio conversational signals at low-latency;
wherein the application server is connected to an agent dashboard operated on the computing device of the agent, the agent dashboard configured to display the computed streamed behavioral signals and objectives for the agent;
wherein the gamification system enables real-time comparison of computed streamed behavioral signals and displayed gamification objectives;
wherein the gamification objectives include one or more target metrics that are identified using machine learning techniques to produce positive outcomes; and
wherein the computed streamed behavioral signals are metrics comprising any of conversational participation computations, conversational flow computations, dynamic variation computations, speaking rate computations, vocal effort computations, and hyper/hypo articulation computations.
2. The system of claim 1 , wherein the application server is connected to an operational datastore configured to store the streamed audio conversational signals and the computed streamed behavioral signals.
3. The system of claim 1 , wherein the application server is connected to a computing device of a supervisor configured to operate a multi-agent dashboard displaying, at the same time, the computed streamed behavioral signals involving the agent and computed streamed behavioral signals involving one or more other agents.
4. The system of claim 1 , wherein the source conversational audio signals are provided to the real-time audio capture system from a private branch exchange.
5. The system of claim 1 , wherein the source conversational audio signals are provided to the real-time audio capture system from a voice over internet protocol switch.
6. The system of claim 1 , wherein the applied metrics are statically configured in advance of receiving the source conversational audio signals.
7. The system of claim 1 , wherein the applied metrics are dynamically configured during a conversation using metrics of a current stream.
8. The system of claim 1 , wherein the applied metrics are configured to analyze one or more non-verbal audio aspects of speech.
9. The system of claim 1 , wherein the gamification objectives include historical metrics.
10. The system of claim 9 , wherein the historical metrics are for prior conversations involving the agent.
11. The system of claim 9 , wherein the historical metrics are for prior conversations involving a team of multiple agents.
12. The system of claim 1 , wherein the gamification objectives include a target value configured at an individual agent, team, or organizational level.
13. The system of claim 12 , wherein the target value is based on learned analysis of connections between computed streamed signals and past business outcomes.