Jev runs my to-do list by voice
Jev is a voice-controlled to-do list application that processes natural language commands to manage tasks. The system uses multiple layers of AI analysis to validate dictation accuracy, match tasks to existing items, and execute the appropriate operation.
Summary
The transcript demonstrates Jev, a voice-activated to-do list management tool that allows users to control their task lists entirely through spoken commands. Users can perform various operations by simply speaking phrases like 'book dentist appointment,' 'remove,' 'buy oat milk,' 'complete,' and 'review pull request low priority.' The application handles the complete workflow of voice command processing through a sophisticated three-layer analysis system. First, it validates the dictation itself to ensure the audio was properly transcribed despite potential background noise or speech patterns. Second, it analyzes whether the transcribed text matches an existing task in the user's to-do list, correcting mismatches when necessary. For example, if a user says 'oat milk' but the system initially transcribes it as something else, it compares against existing tasks to identify the correct match. Third, it maps the recognized task to the appropriate operation to execute, such as 'remove,' 'complete,' or assigning priority levels. This multi-pass approach ensures accuracy throughout the voice command pipeline.
Key Insights
- Jev performs multiple sequential analysis passes on voice input rather than single-pass processing, first validating dictation accuracy, then matching to existing tasks, and finally mapping to appropriate operations
- The system includes error correction logic that recognizes when dictation is imperfect and proactively compares transcribed text against existing tasks to identify matches the user intended
- Voice commands in Jev can specify multiple parameters simultaneously, including the task name, the action to perform (remove, complete), and metadata like priority level in a single natural language statement
- The voice to-do interface achieves task management operations at notably high speed through its multi-layer validation approach, suggesting efficiency comparable to or exceeding traditional text-based apps
- Jev's architecture treats dictation validation, task matching, and operation selection as distinct but interdependent analysis layers rather than a single unified recognition process
Topics
Transcript
[0:00] Traditional to-do app, we have a list of to-dos. I'll just say something like book dentist appointment. Remove. Buy oat milk. Complete. Review pull request. Low priority. Now, the speed there was all done through Jev. It does a few passes of looking through the text where it ensures that it lines up if there's any like dictation mistakes and you say it doesn't get oat milk quite right, it'll go through and see does this quite match up with something or do I need to categorize this into something that exists? It matches it with the thing in [0:31] the list and then it matches it with the function that it's going to call. So, it's multiple layers…
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
What are channels in Claude Duet?
Claude Duet features a channels system inspired by Slack that allows users to organize conversations into separate subthreads within a main chat. This enables more structured collaboration by creating dedicated channels for specific projects, such as a sales channel for sales-related work.
Claude Duet needs multiplayer (please!)
The speaker argues that Claude and similar AI products are currently optimized for single-player use and lack proper multiplayer capabilities. Despite integrations like Claude tags and Slack, the speaker contends that sharing agents with their small team of five people remains problematic, preventing these tools from becoming part of their regular workflow.
Claude Duet is great, but SLOW
The speaker expresses frustration with the slowness of AI model Claude, which impacts their willingness to engage with it. They compare its responsiveness unfavorably against more responsive alternatives, despite acknowledging Claude's capabilities.
Claude Duet should be more annoying
The speaker emphasizes the need for agents like Claude Duet to be more engaging and proactive to avoid being forgotten in the competitive landscape of digital assistants. They critique the current methods of communication, such as morning briefings, as ineffective in attracting user interest.
A perfect TikTok cut from a one sentence prompt
Claude code automatically transforms a single sentence prompt into a polished, TikTok-ready short-form video with captions, jump cuts, and zooms in approximately 4 seconds with no manual editing required. The tool demonstrates AI's capability to understand and execute high-quality short-form video production autonomously.