科大讯飞推出的移动互联网智能交互平台，为开发者免费提供：涵盖语音能力增强型SDK，一站式人机智能语音交互解决方案，专业全面的移动应用分析。
此服务商级交接只保留公开 ID，不会猜测工具或业务任务；请在重新发现时选择当前工具，之后才能把配置视为完整。
Audio moderation API based on intelligent semantic analysis and multilingual recognition technology that accurately identifies sensitive, pornographic, abusive, and other risk audio content.
Long Form ASR that converts long audio data (within 5 hours) into text data based on deep full-sequence convolutional neural networks. Audio files are uploaded and enter a waiting queue, with results available upon successful transcription.
Face watermark photo comparison API that compares a face photo with a watermark photo to determine if they are the same person. If the uploaded photos contain exif orientation information, the service will rotate and flip according to this information before further processing.
Image moderation API based on deep learning technology that accurately identifies various risk image content including prohibited content, politics, pornography, minor violations, violence-terrorism, advertisements, and Logo watermarks.
Perform synchronous compliance check on provided text content using built-in and custom word libraries.
Content moderation API that effectively identifies various risk text content including politics, prohibited content, pornography, violence-terrorism, abuse, and advertisement based on advanced semantic models and massive multilingual sample libraries.
Add a list of words to an existing word library (blacklist or whitelist).
Create a new blacklist word library for use in text moderation.
Add blacklist entries based on lib_id.
Add whitelist entries based on lib_id.
Create a blacklist word library that can be used in text moderation API calls by filling the lib_ids collection value.
Create a whitelist word library that can be used in text moderation API calls by filling the lib_ids collection value.
Delete a blacklist word library based on lib_id.
Delete a whitelist word library based on lib_id.
Video moderation API based on advanced AI technology that comprehensively detects video frames, audio, subtitles, and text to accurately identify various risk video content including sensitive content, pornography, minor violations, violence-terrorism, advertisements, and Logo watermarks.
Submits a list of videos for compliance moderation, including audio and visual content analysis.
Long Form ASR based on deep full-sequence convolutional neural networks, converting long audio data (within 5 hours) into text data.
