Fix: correct speechText for Message_Stream_Event

[sebvanleuven]

Fixing message type for Message_Stream_Event. Fixing message processing, fixing message history

Expand interactive example using MESSAGE_STREAM_EVENTS for live caption and printing the whole message history at the end of a session.

---

Summary by cubic

Adds real-time message streaming support using speechText events and builds automatic conversation history. Fixes message parsing and improves the interactive example with live captions and transcript printing.

• New Features

  • Emit MESSAGE_STREAM_EVENT_RECEIVED with MessageStreamEvent chunks (role, content_index, end_of_speech, interrupted).
  • Track conversation history and emit MESSAGE_HISTORY_UPDATED on completion; added client.get_message_history().
  • MESSAGE_RECEIVED now fires after end_of_speech for backward compatibility.
  • Message now includes id and interrupted; new types exported (MessageStreamEvent).
  • README updated with stream events and talk/talkstream usage; example adds caption toggle (c) and history toggle (h) with full transcript on close.

• Bug Fixes

  • Corrected messageType to speechText and proper role mapping (user/persona).
  • Fixed data channel open-state handling to prevent missed messages.
  • Resolved session close ordering and live caption print order.
  • Minor logging adjustments for clearer output.

^{Written for commit 8229024. Summary will update on new commits.}

[ao-anam]

LGTM.

[cubic-dev-ai]

2 issues found across 7 files

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.


<file name="README.md">

<violation number="1" location="README.md:135">
P2: The example now references `Message` but the import was removed. This makes the snippet invalid when copied. Add `Message` to the import list.</violation>
</file>

<file name="examples/persona_interactive_video.py">

<violation number="1" location="examples/persona_interactive_video.py:41">
P3: The comment says the log level was reduced to WARNING, but the code now sets INFO. This mismatch makes the logging intent unclear for maintainers.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

[cubic-dev-ai]

2 issues found across 7 files

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.


<file name="src/anam/_streaming.py">

<violation number="1" location="src/anam/_streaming.py:345">
P2: Reset _data_channel_open to False when creating a new data channel; otherwise stale True values from a previous session can cause sends while the new channel is still connecting.</violation>
</file>

<file name="examples/persona_interactive_video.py">

<violation number="1" location="examples/persona_interactive_video.py:41">
P3: The comment on the logging level no longer matches the code (INFO vs “reduced to WARNING”), which is misleading for maintenance.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}