conlang

Construct languages from phonetics, to orthography, to grammar, to lexicon, to completion.

The Conlang kit assists in the creation of constructed languages. On a simple level, it can be used as a name, word, or sentence generator: you describe a phoneme inventory and phonotactic patterns in a JSON sketch, and it produces words and sentences in IPA. Beyond that, sketches can describe word classes, morphology (affixation), sentence-template grammars, and lexicons.

What distinguishes this toolkit from others is the focus on phonetics. It is meant to get you from zero to hearing a language spoken as quickly as possible. Generated output is IPA, and a text-to-speech voice driver can speak it aloud, which allows rapid iteration on sounds and patterns until the language sounds like what you want.

Quick start

cargo run --features voice-polly -- config voice init
cargo run --features voice-polly -- generate-sentences --config examples/english.json --speak

This generates sentences from the English-like sketch in examples/english.json, prints them in IPA, and speaks each one aloud through AWS Polly.

To hear audio, you need:

• A voice driver compiled in: --features voice-polly (cloud neural TTS, requires AWS credentials) or --features voice-espeak (free, offline, requires the system eSpeak-ng library).
• A voice configuration: scaffold one with conlang config voice init

Without a voice feature or the --speak flag, generation still works and prints IPA; there is just no audio. Simpler sketches to start from live in examples/.

The CLI

• generate-words: generate words from a sketch (or inline --consonants/--vowels/--pattern flags), optionally restricted to one word class.
• generate-sentences: generate sentences using the sketch's grammar templates, or unstructured strings of words.
• config voice init | list | scan | test: scaffold, inspect, discover, and try out text-to-speech voices.

Pass --speak to the generators to hear the output with a random configured voice, or --speak-with <name> for a specific one.

Documentation

The full documentation gives a deep dive into the fundamentals of spoken language and how this maps to configuration files.

License

Apache-2.0