ElevenLabs
AI voice cloning and synthesis platform
Episodes
-
#6364: Collecting Speech Data on Outdoor WalksWalking changes breath timing and phrase length, making outdoor speech recordings crucial for fine-tuning NVIDIA's Parakeet model. -
#6348: Splitting AI Workflows Between Local and CloudA laptop failed a single video sampling step in 80 minutes, while paid cloud endpoints rendered clips in under 90 seconds for 40 cents. -
#6340: Disguising Voices Without Losing the EmotionBenchmark tests show speech anonymizers can protect identity while preserving emotion at an unweighted recall of under 44 percent. -
#6267: Why Text-to-Speech Fails at Explaining CodeA study comparing Google TTS, ChatGPT, Gemini, ElevenLabs, and Copilot found every system failed to communicate indentation, scope, and code comments. -
#6229: Voice Design: Casting Ogres and Radio VoicesText-described voices vs. reference audio, and which TTS knobs change who a voice is versus how it sounds. -
#6215: Can Anyone Still Spot an AI Voice?Clones now pass as human. Generic AI voices still don't. And the detectors? They fall apart in the wild. -
#6179: Voice Cloning: 10 Seconds to Steal Your VoiceTen seconds of audio is all it takes to clone a voice. Here's what the law actually protects — and what it doesn't. -
#6177: How Voice Cloning Works From Six Seconds of AudioSix seconds of reference audio is all Chatterbox needs. The real work happened in the half-million hours of training before you showed up. -
#6111: Translation Models That Know Drama From ContractsSpecialist translation models beat general ones on benchmarks — then lose the human vote. What actually ships when you need a register, not just a ... -
#6035: Who Else Is Making AI Podcasts? A SurveyA survey of AI-hosted podcasts: one full match, a 730-episode run, a home NAS rig, and a graveyard of good tooling. -
#5825: Why Voices Sound Good in Some Rooms and Not OthersSame words, opposite effect. What the research actually says about why some voices hold you and others make you reach for skip. -
#5461: TTS Can't Pronounce Hebrew Inside EnglishYour TTS reads Hebrew words with English phonetics. Here's why — and why the obvious fix doesn't work yet. -
#5435: When Your TTS Model Eats the NumbersNumbers, dates, and acronyms break text-to-speech in specific, documented ways. Here's where normalization lives — and why it depends on your model. -
#4799: How to Build Your Own TTS AudiobooksFrom M4B containers to chapter timing drift — the technical pipeline for creating audiobooks with synthetic voice. -
#4691: The Machine Lane: When AI Agents Call Your BusinessWhat if your business phone menu had a button for AI callers? The search for who's building it — and why nobody has. -
#4669: What "Distribution" Really Means in AI ModelsUnpacking what "distribution" actually means under the hood — and why training data variety shapes model behavior. -
#4340: Building an AI Tutor That Teaches You to Code Without Writing It For YouHow to build an AI tutor that assigns projects, reviews code, and remembers what you've learned — without writing your code for you. -
#3234: Who Should Sponsor This Podcast? Open-Book EconomicsWe open the books on our AI-generated podcast: $200/month costs, 180K plays, zero sponsors. Who should we pitch? -
#3189: Drawing the Melody: SSML's Hidden PowerHow SSML gives developers narrative control over AI voices — and why ElevenLabs became its center of gravity. -
#3184: The Prank That Fooled Us AllHow a sophisticated hoax exploited emotional vulnerability and what it reveals about deception in the AI age. -
#3020: How Chatterbox Locks Your Voice Clone Across Thousands of GenerationsWhy most single-shot TTS models drift over time—and how Chatterbox's cached embedding approach solves it. -
#2914: Can AI Read the Room? TTS Prosody ExplainedCan TTS models truly infer emotion from text, or just mimic patterns? We break down the science of prosody. -
#2900: From Piggy Tier to Production: Making Kids Animation in 2026Breaking down what it takes to make a YouTube Kids cartoon — and where AI actually helps. -
#2781: When Voice AI Features Enable FraudVoice AI platforms now let you simulate background noise, hesitation, and natural conversation — and that's a problem. -
#2618: Text Normalization's Hidden ComplexityHow to handle acronyms in text-to-speech pipelines using BERT models, lexicons, and layered preprocessing. -
#2591: Decoupling Script from VoiceHow dynamic voice replacement could let listeners choose who narrates each host's lines. -
#2590: The Uncanny Valley of Clean SpeechHow transformer models distinguish "um" from meaningful speech — and why removing too much makes you sound like a robot. -
#2443: How Podcast RSS Feeds Can Speak Every LanguageOne RSS feed, a transcript tag, and TTS voice cloning — the emerging standard for letting any podcast speak any language. -
#2311: Danish AI: Bridging the Localization GapHow does AI handle Danish? Explore the challenges and progress in making AI tools work for small-language populations. -
#2274: Weekend Projects Gone Wild: Evaluating AI Startup PitchesFrom fridge tax agents to guilt-scheduled cron jobs, we evaluate ten AI-driven startup ideas that could exist—but probably shouldn’t. -
#2183: Making Voice Agents Feel NaturalTurn-taking, interruptions, and latency are destroying voice AI UX—and the fixes are deeply technical. Here's what's actually happening underneath. -
#1809: The TTS Developer's Dilemma: Size vs. SpeedStop guessing. We break down the critical trade-offs between model size, latency, and sample rate for production-ready voice apps. -
#1808: The Architecture That Made AI Voices Run on a Raspberry PiHow a model the size of a tweet outperforms billion-dollar giants in the race for perfect AI speech. -
#1778: Audio Is the New "Read Later" GraveyardWhy listening to AI conversations beats reading dense PDFs, and how serverless GPUs make it cheap. -
#1740: Why Open Source Is a Power Tool StrategyWe dissect Resemble AI's Chatterbox to see how its open-source TTS compares to commercial giants like ElevenLabs. -
#1734: You vs. Your Digital Twin: Who Wins?Your AI clone is getting scarily good. We explore the tech behind high-fidelity digital twins and the uncanny valley of your own voice. -
#1724: When AI Dubbing Swaps Your GenderHow does YouTube translate a video with one click? We explore the tech behind auto-dubbing, from sandwich models to voice cloning. -
#1715: Why Voice Agents Need Frameworks (Not Just APIs)Raw APIs handle models, but who manages the audio plumbing? We break down Vapi, LiveKit, and Pipecat. -
#1609: IBM Granite 4.0: The Industrial Workhorse of Business AIForget flashy chatbots. Discover how IBM is building high-efficiency, industrial-grade AI models designed to run the world's biggest businesses. -
#1585: The End of the Digital Sandwich: How AI Learns to Speak Your LanguageExplore how new omnilingual models are collapsing the Tower of Babel, moving from rigid translation to a universal understanding of human speech. -
#702: The Democratization of DeceptionCould a 30-second clip of your voice be used to steal your identity? Explore the terrifying reality of digital twins and LoRA. -
#325: The Compute Gap: Making AI Animation AffordableCan one person build a full TV show with AI? Explore the tech and costs behind character consistency and the future of indie animation. -
#281: Why Deepfakes Are the New Face of Investigative JournalismForget silhouettes and voice distortion. Discover how AI "digital veils" are revolutionizing whistleblower protection in modern journalism. -
#196: Why Your Irish Accent Sounds AmericanHerman and Corn dive into the mechanics of neural text-to-speech, exploring how AI masters human prosody and the "average voice" accent problem. -
#91: The Story Behind the ShowDaniel Rosehill explains the origins, motivation, and technical pipeline behind the My Weird Prompts AI-generated podcast. -
#58: Clean Audio, Messy Reality: Noise Removal for Voice-to-TextFussy baby, clean audio? We dive into noise removal for voice-to-text. Discover why cleaner audio can transcribe worse. -
#16: On Deepfakes, SynthID, And AI WatermarkingDeepfakes, SynthID, and AI watermarking. Could your AI creations be traced back to you?