All AI Pulse updates tagged "TTS".
5 signalsShould you rely on public benchmark scores when choosing a model? @VikParuchuri argues "no" in this talk. To illustrate this, he shows specific problems found in popular OCR benchmarks: ParseBench: • A test expects 310 gigawatts of solar-panel capacity when th
FishAudio releases Drama 3, an open-weight TTS model with word-by-word control of emotion, tone, and delivery. Build multi-speaker dialogue and voice agents. API usage is now 80% off.
The poster says FishAudio released open-weight Drama 3, an emotive controllable TTS model that avoids regenerating a whole clip when fixing one bad word.
Soniox TTS v2 is one of those models you need to actually hear. They just launched TTS v2, a text-to-speech model that speaks 60+ languages from one mode. Really premium voice quality at a dramatically lower price ($ 0.70-per-generated-hour). > One model repla
I didn't believe this was real; the video shows incredible detail.