Spotify is reportedly developing AI tech that can clone podcast hosts' voices to generate ads without them ever stepping into a booth. The tool would let creators produce localized or multi-language sponsor reads with their permission. Spotify hasn't officially confirmed the feature, but admitted it's actively exploring generative audio.
Meta just open-sourced its Massively Multilingual Speech (MMS) models, expanding text-to-speech and speech-to-text capabilities to over 1,100 languages—a tenfold leap over current tools. It can also identify more than 4,000 spoken languages, trained on diverse recordings of translated religious texts. It is a critical push toward saving endangered dialects and bringing speech AI to underserved communities.
Apple just previewed a new slate of accessibility features, headlined by Personal Voice, a tool that creates a personal speech twin using 15 minutes of training audio. Designed for people at risk of losing their speaking ability, the model processes entirely on-device to keep voice data private. The features arrive later this year.
Wendy's just teamed up with Google to launch 'Wendy's FreshAI,' an automated chatbot built to take drive-thru orders. Trained on internal slang like 'JBC,' the bot begins pilot testing at a Columbus, Ohio location in June with human workers monitoring the line.
Humane co-founder Imran Chaudhri just gave the world its first full look at the company’s secretive screenless wearable in a TED talk. The clip-on device operates independently of a phone, projecting calls onto your palm, summarizing emails, and translating speech using an AI voice clone. Why it matters: Humane wants to replace smartphone screens entirely, though how practical this puck is in everyday life remains an open question.
Google just partnered with implant maker Cochlear and hearing research groups to develop personalized hearing devices using AI. The technology isolates and separates specific voices from background chaos, making conversations much easier to follow in crowded restaurants and transit.
Grimes just launched Elf.Tech, a new AI voice software that lets creators transform vocal recordings into her signature sound. Anyone can use the generated vocals to produce and distribute new tracks, provided they agree to a 50% split on master recording royalties. Why it matters: Rather than policing AI copies, she is turning voice duplication into an open-source revenue share.
Amazon Music is getting an AI upgrade to automatically generate podcast episode highlights and chapter markers. The feature stems from Amazon's acquisition of audio startup Snackable AI, allowing listeners to jump directly to specific segments without having to scrub through full episodes.
ElevenLabs just dropped Eleven Multilingual v1, bringing its expressive text-to-speech tech to seven new languages, including French, German, Hindi, and Spanish. The model preserves a speaker's unique tone and accent across languages, working seamlessly with existing voice-cloning tools. Why it matters: Localizing games, audiobooks, and media just got much easier, though non-English numbers and acronyms still occasionally trip it up.
Grimes just invited creators to clone her voice with AI, promising a 50% royalty split on any hit track. Free from record label contracts, she announced she wants to "open source all art" and is happy to be a guinea pig, taking a completely different stance from industry crackdowns.
Humane co-founder Imran Chaudhri just gave a sneak peek of the startup’s standalone AI wearable at TED. The screenless gadget operates without a smartphone, uses voice and gestures, and projects caller info directly onto your palm in green light. Chaudhri also demoed real-time French translation, with more details promised in the coming months.
Scammers in Texas just used AI voice cloning to impersonate a couple's son, fabricating a hit-and-run crash to swindle $5,000 in fake bail money. Experts warn criminals can clone convincing voices using short public audio clips from social media. It serves as a stark reminder to establish family safe words and verify emergencies.
Google just shared new details on its Universal Speech Model (USM), a 2-billion-parameter system trained on 12 million hours of audio across 300+ languages. It currently supports speech recognition and translation for over 100 languages, and is already powering YouTube's closed captions. Why it matters: It marks Google's first critical milestone toward understanding the world's 1,000 most-spoken languages.
OpenAI just made GPT-3.5 Turbo and Whisper available via API, slashing ChatGPT costs by 90% down to $0.002 per 1,000 tokens. Whisper speech-to-text runs at $0.006 per minute, and OpenAI will no longer train on developers' API data by default.
Voicemod just raised $14.5 million to expand beyond gaming soundboards into real-time generative AI voice synthesis. Fresh off integrating tech from its Voctro Labs acquisition, the startup is developing tools that let users sing in someone else's voice or swap vocal identities on calls. Why it matters: Much like viral camera filters changed video, real-time AI avatars are turning personal voice identity into something anyone can customize on the fly.
Spotify just rolled out DJ, a new beta feature that pairs personalized music recommendations with an uncannily realistic voice commentary. Powered by OpenAI's tech and Sonantic's voice engine, it gives context on artists and songs as it plays. If the vibe isn't right, users can simply tap the DJ button to switch genres instantly.
Apple and audiobook distributor Findaway have agreed to halt using audio files for machine learning training following pushback from voice actors and union SAG-AFTRA. Narrators discovered a buried contract clause letting Apple feed their voices into AI models designed to automate narration. While file access is now paused for union members, discussions over compensation and data safeguards are ongoing.
TikTok creators are sharing hilarious and exasperating run-ins with McDonald's voice-ordering bots, showing the AI mistaking vanilla ice cream for butter packets and running up a $250 McNugget tab. Fast-food chains have chased automation to ease labor shortages, but the bot's stubborn confusion proves human drive-thru crews aren't getting replaced anytime soon.
Amazon CEO Andy Jassy acknowledged the excitement around generative AI like ChatGPT, but insisted the tech giant has been building its own large models for a long time. Amid investor concerns that Microsoft has stolen the spotlight, Amazon is exploring partnerships and working to make Alexa far more conversational.
A Twitch channel called Nothing, Forever has been broadcasting an endless, AI-generated Seinfeld parody for over a month. Powered by GPT-3 and generative algorithms, the system scripts dialogue, voices lines, and directs camera angles in real time. It is glitchy and deeply surreal, but it offers a fascinating look at the future of infinite, automated entertainment.
