ElevenLabs shows a complete workflow for producing an ad with a generated jingle, from script to final audio.

Official Higgsfield walkthrough of driving video edits with Claude Opus 5.5 and the Katana tool, an agent-plus-video workflow.

LiveKit's Jesse Hall treats voice-agent latency as a budget and shows how to make conversations feel humanlike.

A workflow trick from Curious Refuge for cutting the cost of AI video generations, practical for anyone iterating on video.

AssemblyAI's workshop compares async, sync, and realtime speech models so you can pick the right one for your voice app.

A behind-the-scenes look at the real workflow and pitfalls of making a 17-minute AI-generated film.

A full step-by-step workflow for making seamless AI video intros, with the process shown rather than just described.

Full tutorial on producing a cinematic AI commercial with Higgsfield Cinema Studio and Seedance 2.5, with every prompt and asset provided.

Google Cloud Tech shows a design pattern for live voice agents that drive a real browser with the Gemini Live API while keeping the conversation going.

Kevin Stratvert shows how to edit videos with ChatGPT and Codex, turning rough footage into cuts, captions and exports without a traditional editor.

Higgsfield's DevDay 2026 session on building agents that provision their own compute, from a team shipping AI video at scale on OpenAI's platform.

Official ElevenLabs launch of ElevenAgents Architect, an in-product expert that analyzes your voice-agent transcripts, proposes fixes and validates them in simulation…

Theoretically Media tests two new open-source video models (Prism with native 2K plus audio, Kandinsky 6.0 on a 4090), Nano Banana 2.1, and mods DOOM into Skyrim with a…

Arize on debugging voice agents where the transcript looks fine but the call failed, and the tracing and evals you need for speech-to-speech systems.

Higgsfield pits its AI video tools against a professional VFX artist on the same shots, a useful reality check on where AI video is good enough for production.

Suno's official walkthrough of Albums: creating, organizing and publishing a full album of AI-generated music from your Suno library.

IBM's explainer on how generative music models work, how Suno-style tools are changing creation and licensing, and what it means for builders in audio.

Higgsfield shows a full GPT-6 Astra-driven AI video editing workflow, from prompt to cut, with the practical steps builders can copy today.

Tech With Tim builds a real-time conversational AI avatar end to end: speech-to-text, LLM, text-to-speech and a lip-synced video avatar, with full code.

Suno's own tutorial for its new Speech beta, covering spoken-word generation, voice styling and mixing speech into generated tracks.

Theoretically Media rounds up the week's generative media releases in one pass: Kling 4, Gemini 4's video and image updates, ElevenLabs v4 and Ideogram 4.5, with real…

Hannah Fry with DeepMind's Pushmeet Kohli on how SynthID watermarks AI text, images and video, and how it now extends to protein sequences.

PolyAI CTO Shawn Wen explains why voice agents are harder than text agents and how an audio-native model predicts turn-taking before it replies.

Runway Labs introduces Continuum, a real-time video operating system that generates and edits video live, a glimpse of where AI video tooling goes next.

Runway's official launch of Runway Ads, a generative video product aimed squarely at producing ad creative end to end.

Runway shows how to build an automated ad-creative pipeline on the Runway Dev API, generating and iterating video variants programmatically.

ElevenLabs' official rundown of Eleven v4 and v4 Turbo: new voice quality, latency for agents, languages and what changes for existing integrations.

The official ElevenLabs launch of Eleven v4 and v4 Turbo, its new flagship speech models for expressive TTS and low-latency voice agents.

Claude plans the script and shot list, Higgsfield renders it: an automated pipeline for short-form video that a one-person team can run daily.

Text-to-speech, voice cloning, dubbing and the studio in eleven minutes. Current as of mid-2026.

A filmmaker's walkthrough of Higgsfield's cinematic controls and how to get consistent characters and shots across a sequence.

A ten-minute production workflow for ads and product videos in Higgsfield, including prompt structure and export settings.

Higgsfield from zero: models, camera-motion presets, image-to-video, and a full short-form video produced on screen.

ElevenLabs' official guide to a conversational voice agent: voice, knowledge base, tools, phone number and deployment. The reference for anyone adding voice to a product.