Speech: Buddy's SpeechCenter
Buddy's SpeechCenter is an easy way to add voice interactions to robots that run under Microsoft Windows. With the SpeechCenter, you can turn your 'bot into a chatbot.
Unlike other chatbot software, which are designed for text input, the SpeechCenter is designed to use speech recognition as input, and is intended for spoken interactions.
The replies do not need to be fixed strings, but can instead use randomly selected variations in phrasing. For example, Buddy might answer the question, "What's your favorite color?" with
Predefined SpeechCenter replies can also include audio files (which will be played), or text files (which will be read aloud). This allows your robot to respond, like R2D2, with expressive sounds instead of words, or use prerecorded phrases. Text files can be used to organize longer responses, such as factoids, or can be used to store dynamic information that your program updates, such as the current weather forecast.
The recognition and speaking portions can share definitions, which can range from simple, hardcoded strings, to complex, multi-level, nested definitions that include multiple patterns made up of lists of equivalent clauses. During recognition, any of the possible variations results in a recognition, and when speaking, the variations for each level are chosen at random.
You can also insert wildcards that will allow the SpeechCenter to recognize speech that's not quite an exact match.
In addition to predefined chatbot interactions, which you define in one or more files, the SpeechCenter also includes interfaces for voice output (including speaking, spelling, playing audio files, and reading text files) that your own programs can use directly.
The SpeechCenter uses Microsoft's Speech API (SAPI).
Download Buddy's SpeechCenter