Speechmatics-flow

Changed

- Removed `choices` from `--assistant` cli argument to support arbitrary strings
- Implemented case-insensitive lookup for predefined template names

Changed

- Include the package version when requesting a JWT

Added

- Added Debug mode
- Introduced a new debug field in `StartConversation`.
- When LLM debug mode is enabled, the client will receive debug messages with LLM request and response content

Fixed

- Issue with Flow latency on Linux

Changed

- Examples load env vars from .env file
- Examples now print user/agent transcripts

Added

- Introduced a new class to enable custom configuration of audio playback settings, offering greater flexibility for
fine-tuning audio playback.
- The client now buffers audio to ensure smoother playback, especially in challenging network conditions.

Fixed

- Resolved an issue with reading piped audio from stdin.

Added

- Client-side function calling example
- Improved audio playback in examples
- Added instructions on how to generate an API KEY and how to install pyaudio

Changed

- Set default URL when running from CLI