Oral News Broadcast Script is optimized for spoken delivery, rewriting written news material into a script tailored for radio anchors, TV news anchors, podcast hosts, or self-media video presenters. Sentences are kept short, natural pauses are inserted, awkward long phrases and obscure characters are avoided so that voice talent can deliver fluently in one take. The script auto-prepends a familiar opening (e.g. "Good evening listeners") and a sign-off (e.g. "That's all for this segment, thanks for tuning in"), and rewrites numbers into easily-spoken forms ("six thousand two hundred" instead of "6,200") to avoid tongue-twisters like "thirty-two point seven percent". Runs entirely on local, so iterate to fit your show length. The output is ready for direct delivery, can be paired with text_to_speech_byteplus for AI voiceover, or dropped into veo3_1 / cheer_video / christmas_video to produce a complete video news package. For written press-release format, use news_report instead. Recommended for radio anchors, news podcasters, self-media news channels, and any internal communication that needs spoken-word style. Maximum context window (input and output combined): about 1,048,576 tokens. Maximum output per reply: about 65,536 tokens (model reasoning, if any, counts toward this limit). Maximum total upload size per request: 256MB (all files and message content combined). Accepted document formats: pdf, docx, xlsx, pptx, txt, csv (file content is provided to the model as extracted text). Accepted image formats: heic, heif, jpeg, jpg, png, webp. Maximum number of images: 3600. Maximum size per image: 14MB. Maximum total image size per message: 14MB (text and history share the same request size limit; uploads close to the limit may still be rejected). Image dimensions: at least 200 pixels per side. Maximum image resolution: 36,000,000 pixels in total. Output content formats: Text This module supports continuous conversations. AI can make mistakes. Please verify important information.