deepgram_speech_to_text

• fix keyterms serialization (thanks @agladyshev)
• add Finalize method (thanks @agladyshev)

• improve models (thanks @jbienzss)
• cleaner error handling in websockets (handle errors in the stream)

• avoid JSON parsing multiple times in DeepgramListenResult

• Breaking change: Remove baseParams from DeepgramClient since it creates confusion.
• Support using short-lived JWT tokens instead of API KEY (thanks @danielmahon)

• allow custom baseUrl for Deepgram Client

• fix utf8 encoding

• removed unnecessary exports

• Updated docs

• Updated docs
• added missing exports

Breaking changes

Now you can find all method under deepgram.listen and deepgram.speak

• deepgram.transcribeFromFile() -> deepgram.listen.file()
• deepgram.speakFromText() -> deepgram.speak.text()
• Added support for deepgram.speak.live() - deepgram.toWav() makes the audio data readable
• Updated examples (flutter and dart)
• Added mandatory default query params to every streaming method (encoding, sampleRate)

• Stop keepalive timer on close (thanks @PcolBP)

• Updated docs

• Added pause() and resume() methods for streaming
• Add missing streaming features (thanks @DamienDeepgram)
• Fixed: Memory leak + infinity future (thanks @PcolBP)
• DeepgramSttResult now has .type getter, .transcript is safer and nullable

• Fixed utf8 parsing issue

• Update web_socket_channel dependency

• Added transcribeFromPath() for convenience.
• Fixed package on web.
• Updated example to work on web.

• Support for TTS (text to speech)

• Better documentation and debug hints

• STT methods now return a DeepgramSttResult instead of plain json string (breaking change)
• Updated example in README.md and /example, added a flutter demo
• Better documentation

• Better doc

• Added WEB support (for real this time)

• Tried to add WEB support

• Fixed incorrect documentation

• Formatted for pub.dev

• Initial version.