CHANGELOG

## [2.1] - 3 Jun 2021

### Added

* download-profile command and auto-download of relevant language files from Github
* New Kaldi profiles for Czech, English, Spanish, French, Dutch, Russian, and Swedish
* New DeepSpeech profiles for German, French, and Italian

### Changed

* Upgrade DeepSpeech support to v0.9
* --profile argument can now be a language name, profile name, or directory

## [2.0] - 7 May 2020

### Added

* Support for Mozilla's DeepSpeech speech to text engine (version 0.6.1).
* transcribe-stream for live transcription
* Number ranges, slot programs, and converters to template language.
* --intent-filter for recognize-intent command.
* --certfile and --keyfile arguments for SSL support.
* print-downloads command (download only necessary files).
* print-files command (for backups).
* print-version command (or --version).

### Changed

* wait-wake command uses Mycroft Precise instead of porcupine.
* speak-sentence command takes --marytts flag instead of --espeak. Defaults to eSpeak.
* Using autoconf for source install.
* Minimum required Python version is 3.7

### Removed

* MaryTTS server is no longer embedded in Docker image. Must be hosted externally.

## [1.0] - 20 November 2019

* First version