#语音

Parakeet Ultra from @moondreamai is live on OpenRouter Moondream's post-trained version of NVIDIA Parakeet TDT 0.6B v3: same architecture and speed, lower word error rate on every benchmark group. 25 European languages, with segment and word timestamps. $0.06 per hour of audio, served by @PrimeIntellect. http://openrouter.ai/moondream/parakeet-ultra

HeyGen Voice Turbo from @HeyGen is free on OpenRouter until November 30! The lower-latency version of HeyGen's text-to-speech model, the one that topped the Artificial Analysis TTS leaderboard. $0, no credits needed. Over 2,900 voice options, 39 languages, up to 5,000 characters per request, streamed as 44.1 kHz audio. See more: http://openrouter.ai/heygen/heygen-voice-1-turbo:free

Introducing the Hume Audio Decision API. Ask anything: How many speakers? Is anyone laughing? Is the caller frustrated? What used to take 10+ models is now one call. Trained on 10M+ human judgments, so it answers the way human listeners would. Realtime, or across hours of recordings. Request access below.

Tested voice cloning with Ming-3.0-tts-0.5B by @AntLingAGI and it is even more insane. For a model of this size, voice cloning happens really well, i tested my own voice, half drunk, in a room full of noise (fan noise mostly) and it instantly cloned it. It needed just a 5-second sample of my voice, and it speaks faster than real time: about 4 seconds of speech in 2–3 seconds. All running locally on my RTX 3090. Ming takes about 3.3 GB of VRAM, and the whole model is under 3 GB on disk. Great time for Local AI.

Drop 2/15 Are we actually measuring the things we ultimately want our voice AI models to achieve? The best AI voices are becoming hard to tell apart. From each other. Not from humans. 🚀 Introducing Turing TTS: an evaluation that measures how often human listeners mistake AI-generated speech for real human speech. We tested leading TTS models against real human speech. Even the best model fooled listeners only ~10% of the time. With Turing TTS, we hope to contribute to a broader discussion within the research community about how we measure progress in voice AI. If one of our goals is to develop speech that is indistinguishable from human speech, how well do our current evaluation methods capture progress toward that goal? We don’t claim that Turing TTS is the right way forward, or that human indistinguishability should be the goal of every TTS system. Rather, we hope these findings encourage the community to collectively examine whether blind pairwise human evaluations of naturalness are sufficient, and what else we might need to measure. Results are now live on @voicearena_ai.

Introducing Cartesia Managed Agents: our end-to-end voice agent platform for enterprise. #1 on @ArtificialAnlys Speech Agent Arena and @ServiceNow EVA-Bench, built by co-designing our voice models and harness as one system. Check it out → https://cartesia.ai/blog/managed-agents

Granola now does voice typing. Unlimited words, for free. Granola already writes your meeting notes. Now it can do the typing for you, in any app.

Drop 1/15: Introducing Joy, a dual-channel conversational dataset. Real, unscripted conversations between two people, each on their own track. Built to teach voice models how people actually talk: when to listen, when to jump in, when to say “hmm”. 100,000 hours. 50,000 speakers. 10 languages.

Introducing Phonic-1, the most advanced speech-to-speech model for agentic tasks — from @phonic_co, the speech-to-speech model company. We’ve had a breakout year and raised a $27.5M Series A led by @firstround with participation from @Lux_Capital, @felicis, and Definition.

Live Chat is now 100% rolled out to all Ultra users on mobile 📱✨ Have real-time voice conversations with your notebooks across ~100 languages. Powered by our latest audio models, ask questions about your sources and get step-by-step guidance (almost) completely hands-free.

We worked with @therealbradysm, @annuskabarta, and the rest of the multimodal team to bring this incredible new voice mode to Perplexity web. - Natural back-and-forth conversation - Navigate the app entirely by voice - Work across multiple sessions at once Runs on GPT-Live-1. Lots more improvements to come!