Television Academy's new AI guidelines: A step forward for audio description

🎙️ New Experiment - “1940s German Broadcast Voice - Cinematic Edition” I recently ran another experiment with one of my new Professional Voice Clones (PVCs) at ElevenLabs. This time, I wanted to explore a different angle: What happens when you don’t just recreate a voice - but an entire era? The goal was to simulate the tone, posture, and spirit of a 1940s broadcast announcer - the kind you'd hear before a newsreel or cinematic feature. But instead of simply imitating, I fed stylistic input directly into the model - tone, pacing, pronunciation - and let ElevenLabs do the rest. The result: A German-language voice model that doesn’t just speak like 1940… it feels like 1940. Warm, linear, and unmistakably vintage. The voice was originally recorded in native German, but this video uses Eleven Multilingual v2 to render it in English – without losing the era-specific tone. 🎧 Voice was recorded using a retro-style guitar preamp to shape the signal chain. (A fun discovery: running your voice through a guitar amp doesn't break the model - it makes it time travel.) 🎞️ The accompanying video imagines the world of 2025… as seen from the 1940s. (Yes, with typewriters, cigarette boxes, and "moving pictures on demand.") 🧪 And the real insight? Stylistic conditioning works. Even nuance and cadence can be translated into synthetic speech - reliably. Here’s the model: “1940s German Broadcast Voice – Cinematic Edition” https://lnkd.in/eVcv-PUT Curious to hear your thoughts. Do you think style transfer could become the next frontier in voice AI? #AIvoice #ElevenLabs #VoiceClone #PVC #1940s #BroadcastVoice #VintageAI #RetroVoice #SpeechSynthesis #VoiceDesign #GermanVoice #CreativeAI #SyntheticVoice #HistoricalSound #RetroSound

Why Human Voice Performance Still Matters: Channelling Intensity in a Satirical Context - AUDIO ON! 🔊 This piece features my performance of the famous bunker scene from Downfall - reimagined as a commentary on global food policy. After years in voice performance, what interests me most isn't whether Generative AI can mimic voices, but how human artistry creates the subtle emotional authenticity that makes satirical content both funny and compelling. #VoiceActor #Drama #ClassicCinema #WeeklyVoiceSeries

One of the biggest barriers between creators and their audiences is surprisingly simple: unclear dialogue. According to several studies, over half of viewers now turn on subtitles even when watching in their own language. There are a range of reasons, including a distracting background mix, the audio quality & capabilities of the playback device, or even age-related hearing loss. But when you miss a line in a show or movie, you don’t just lose the words - you lose the emotion and the nuance to the story. That’s where I see AI playing a quiet but important role. There’s still a lot of necessary debate to be had in using it to create content, but I think there’s a lot of opportunity in using it to support content delivery - to make the dialogue more audible, more intelligible, and more aligned with what the creator intended. These features can now adjust in real time, even allowing users to choose the level of enhancement that’s right for them, offering perfect (and personalized) audio. Helping people experience a story fully is one of the best ways we can support the people who create it - and AI can do just that.

#VibeVoice is designed for generating expressive, long-form, #multispeaker #conversational #audio, such as podcasts, from text. It addresses significant challenges in traditional Text-to-Speech (#TTS) systems, particularly in scalability, speaker consistency, and natural turn-taking. A core innovation of VibeVoice is its use of #continuous #speech #tokenizers (Acoustic and Semantic) operating at an ultra-low frame rate of 7.5 Hz. VibeVoice employs a next-token #diffusion framework, leveraging a Large Language Model (#LLM) to understand #textual #context and #dialogue flow, and a diffusion head to generate high-fidelity acoustic details. The model can synthesize speech up to 90 minutes long with up to 4 distinct speakers, surpassing the typical 1-2 speaker limits of many prior models #openweights https://lnkd.in/giz3tB_D

YouTube announced new generative AI for Shorts: Veo 3 Fast text-to-video (480p, lower latency) with sound, rolling out in the US, UK, Canada, Australia, and New Zealand. Coming soon: motion transfer to animate images, style filters, and text-added objects. A Speech to Song remix uses Lyria 2 to turn dialogue into tracks with vibes. Edit with AI builds first drafts, adds music, transitions, and reactive voiceover (English/Hindi). Tests expand in the coming weeks.

AudioCipher Technologies Ezra Sandzer-Bell breaks down how serious producers are actually building creative tech stacks with Suno, Udio, and ElevenLabs. The future isn't in pop generation—it's in multimedia storytelling. Film scoring offers creative freedom across multiple styles within single projects, exactly what AI tools enable. Most striking quote: "I've been in dozens of cafes running 12-hour Suno playlists. It's functional music with no care given to sonic branding—this is where AI music risks poisoning public spaces." For creative professionals exploring AI workflows, this conversation cuts through the hype to reveal practical techniques like cross-pollination between platforms and sophisticated prompt engineering. Essential listening for anyone building creative workflows with AI tools. Thanks Ezra! Link to full episode in the comments

When people hear “AI video,” they often picture text prompts generating clips. But real-time AI video is much richer. Depending on the tool, you can: • Transform your live camera feed with prompts • Use an image as a reference for style or texture • Run transformations on an existing video stream • Experiment with audio-reactive visuals tied to music or voice This variety of inputs is what makes real-time AI video such a powerful creative medium — it doesn’t just generate content, it responds to you. Visuals generated with http://jo.my/zyfvdx

🎙️✨ Voice Design Magic: Create the Perfect Voice for Your Project Imagine needing the perfect voice but not finding exactly what you want in the library... What do you do? That’s where the power of Voice Design with Eleven Labs comes in. This tool lets you create unique voices from scratch, tailored to your vision. Want a young voice with a thick French accent and a slow pace? You can make it happen! 🧙♂️🔊 The secret is in the prompt: the more details you give—age, tone, accent, pacing, audio quality—the closer the voice will be to what you imagine. And the best part, you can set it from crystal-clear studio quality to low-fidelity for calls or old recordings. 🎧 Also, the text you use to preview the voice matters as much as the description. Want it to sound natural and coherent? Give the model a good script to deliver magic in every word. ✨ Have you tried designing voices with AI? What’s been your biggest challenge or discovery? Comment and share your story; we all learn together here. 👇 #ElevenLabs #VoiceDesign #Innovation #VoiceTech #AI #Creativity

A great AI voice isn't just about picking the right TTS model. It's the combination of voice quality and voice experience. In our latest video, Meghan Berton breaks it down into two pillars: 1. The quality of the synthetic voice itself (model selection, fine-tuning, expressiveness) 2. The full voice user experience (low-latency infrastructure, audio management, and empathetic dialogue design) All of it comes together inside Agent Studio. Watch the full video here: https://lnkd.in/eDaJc8Ex

Runway: Creators now have the power to swap voices, even after the video is done. This is virtual dubbing, unlocked. Here’s what you can now do in Runway Act-2: → Upload a video and generate a new voice → Apply voice changes to existing or generated videos → Add variety and realism to performances → Alter voices post-production This feature is incredible for indie cinema, content creation, and anyone chasing creative freedom. Lost with AI Video production tools? ────── ✦ ────── ✅ 𝘽𝙚𝙨𝙩 𝙂𝙚𝙣 𝘼𝙄 𝙏𝙤𝙤𝙡𝙨 & 𝙒𝙤𝙧𝙠𝙛𝙡𝙤𝙬𝙨 𝙀𝙭𝙥𝙡𝙖𝙞𝙣𝙚𝙙 (𝙁𝙧𝙚𝙚 LIVE 𝙒𝙤𝙧𝙠𝙨𝙝𝙤𝙥) My personal AI pipeline and tips for effective AI Video production ⬇️ https://lnkd.in/d747VzTe