OpenAI’s latest moves put many voice AI startups on notice

Voice AI weekly digest

Top Updates 💪

• OpenAI’s latest moves put many voice AI startups on notice (CX Today)

• Apple plans AI search engine for Siri to rival OpenAI, Perplexity (Bloomberg)

• This new tech could save old languages from dying out (CNN)

• China’s AI recorder from Alibaba’s DingTalk (South China Morning Post)

• Taco Bell is adjusting its Voice AI plans (Yahoo Finance)

• Meet Whispering, an open-source, local-first transcription app (Slator)

• Tribal Affairs Ministry of India unveils AI translator Adi Vaani (MobileAppDaily)

• Why voice still rules in the AI-powered contact center (TechRadar Pro)

• HappyRobot raises $44M to scale AI agents for freight operators (Reuters)

• Recall.ai is unlocking meeting data to power AI applications and agents (BVP)

• Intella raises $12.5M to scale Arabic speech intelligence platform (SiliconAngle)

• Mega raises $2M for order-to-cash management with AI agents (Financial IT)

• Supersonik secures $5M from Andreessen Horowitz for its AI agent (CityBiz)

• Moonshine AI is developing on-device voice AI offline models (Tech in Asia)

• Synthesias AI clones are more expressive than ever (Technology Review)

• NotebookLM adds Audio Overview AI for analysis and debate (Android Central)

• Jio Haptik launches AI agents for small businesses (The Hindu Business Line)

• Hyro launches Proactive Px™ with AI agent (PR Newswire)

• Cerence AI launches AI Agent for cars to enable safer, smarter work on the go (Cerence)

• From the bakery to the warehouse: BAKO adopts pick-by-voice (PresseBox)

• Waanee AI launches bidirectional multilingual voice translation (EIN Presswire)

• Factors driving voice AI adoption in industry (TechBullion)

• Honor introduces on-device speech recognition and real-time translation model supporting Arabic (Emirates24|7)

Voice AI Podcast 🎙️

In case you missed the latest episode of Voice AI Podcast…

---

Engineering Corner 😎

• Dymesty AI Glasses: Titanium eyewear full AI assistant (Android Authority)

• TaDiCodec speech tokenizer delivers extreme compression with high quality (X)

• Chatterbox Multilingual: An open-source zero-shot TTS model with emotion control and watermarking (MarkTechPost)

• Pipecat: Real-time voice & multimodal AI agents (PyPI)

• LoRA-INT8 Whisper: a low-cost Cantonese ASR for edge devices (Sensors)

• Super WhatsApp Agent: multi-modal assistant in WhatsApp (Dev.to)

• Speech-to-speech translation pipeline with voice-cloning and lip-sync (GitHub)

• Assessing speech recognition in Persian-speaking children (Scienmag)

• Build AI STT and TTS accessibility tools with Python (freeCodeCamp)

• SpeakerMatch: Matching reliable pseudo-labels in semi-supervised and self-supervised speaker recognition with confidence distribution (ScienceDirect)

• Kyutai vs Whisper: Streaming STT AI models compared (Geeky Gadgets)

• Dreamface Voice Clone: The most realistic AI voice generator for YouTubers, businesses, and creators in 2025 (AI Invest)

• Best free TTS software for 2025 (Business.com)