TokenLean·11h agoMurf vs ElevenLabs API: what 1 million characters of speech really costs (Nov 2026)Murf vs ElevenLabs API: what 1 million characters of speech really costs (Nov 2026)
Alex Wu·1d agoWriting for an AI Kid Voice: A Short Scene, RewrittenA paper boat, a small disagreement, and a script you can actually use.
Henry Yan·1d agoShengChang: Import a Novel Chapter, Get a Mixed Radio-Drama EpisodeText-to-speech is basically solved for reading text out loud. Turning a story into something that sounds like a radio drama is not. Once…
Hwnuist·2d agoHow I Clone My Own Voice on a Laptop With No GPU (Qwen3-TTS)The CPU-only setup that worked, the setting that made my clone sound choppy, and how I pick the best take.
InOpen IntelligencebyDr. Fadi Shaar·3d agoVoiceStudio: A Complete AI Voice Studio Runs on Your Own HardwareClone voices, design new ones, dub videos, transcribe speech, and generate narration locally, with an open-source application supporting…
Gabor Gulyas·4d agoHow I Turned My Math Article Into a 14-Minute Animated VideoA while ago I wrote The Hidden Geometry of Numbers: every whole number is a point in a lattice with one axis per prime, multiplying is…A response icon2A response icon2
Oğuzhan Hızıroğlu·5d agoConverting Audio Recordings into Multilingual Text with Amazon Transcribe and Amazon TranslateHello everyone. I’m excited to share a new blog post with you. This time, we’re diving into the world of audio by chaining together two AWS…
InOpen IntelligencebyDr. Fadi Shaar·5d agoHigh-Quality AI Voices Without a GPU: Meet KittenTTSHow an open-source TTS project brings compact speech models, voice cloning, expressive controls, and local audio generation to everyday…
InJIN System ArchitectbyJIN·5d agoThe Four Things You Hand Over When Speech Moves onto the Device: A Teardown of Gradium PhononPhonon is a 31M-parameter text-to-speech (TTS) model with a 64.5 MB weight file. The model isn’t the interesting part. What matters is…
Maks Khomutskyi·6d agoLocal Speech on Intel Arc: Run Kokoro TTS and Whisper STTA practical B50 deployment with PyTorch XPU, OpenVINO GenAI and two Docker workers.
