Voice Tools

• Voicebox is an open-source, fully local TTS tool powered by Alibaba's Qwen3-TTS that achieves near-perfect voice cloning without any cloud dependency (from voicebox local tts open source)
• Voicebox includes a DAW-like "Stories Editor" for composing and editing voice output, making it a production-ready tool rather than just a model wrapper (from voicebox local tts open source)
• Local open-source TTS at this quality level directly threatens ElevenLabs' paid cloud API model -- signals commoditization of voice synthesis (from voicebox local tts open source)

Multi-Modal Personal Agent Requirements

• Peter Yang's framework explicitly includes mid-conversation modality switching (text → voice → video → live calling) as a requirement for "great" personal agents — voice can no longer be a separate product surface; it has to be one fluid switch inside the same agent session (from personal agent requirements framework)

Real-Time Meeting Transcription

• Codex Meeting Recorder skill uses GPT Realtime Whisper endpoint for live transcription at $0.017 per minute ($0.51 for 30-minute meetings), displaying live transcription in preview pane — allows asking questions about transcript content as it's being generated (from codex realtime meeting transcription gpt whisper)
• A local realtime transcription option using Nemotron Speech Streaming is being considered as a cost-effective alternative to GPT Realtime Whisper for meeting recording (from codex realtime meeting transcription gpt whisper)

Voice Input for Agent Workflows

• Speech-to-text apps like Monologue are recommended for communicating with coding agents (Codex) more efficiently, especially for repeated workflows — dictation is faster than typing for the iterative instruction loop that agentic development requires (from gpt codex frontend prototype workflow)

voice-orchestration

• Workflow: open ChatGPT Voice, ask for status + recommended next step on all current projects, approve recommendations, brain-dump on one project, iterate until an idea forms, then say 'start a thread on that' and 'what's next' — repeat across projects. (from chatgpt voice orchestrator workflow)
• Core principle: treat Voice as a chief of staff you ask questions of (more questions than commands), not a worker you command directly — Voice itself doesn't do the work, it hands off good ideas to dedicated agents/threads on your main machine. (from chatgpt voice orchestrator workflow)
• Setup tips: keep a 'compass' doc per project listing short/long-term goals so Voice can recommend next steps from it; designate one device as HQ and use others as remote nodes into it (e.g., ChatGPT iOS remoting into a Mac Studio during walks). (from chatgpt voice orchestrator workflow)
• Claimed productivity effect: this 3-step cycle ('where do we stand' → 'what's next' → 'spin up a thread') made the author 10x more productive over 2 days versus prior workflows, attributed to lower cognitive/energy load from delegating brainstorming to the AI. (from chatgpt voice orchestrator workflow)