wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[WIP] AWS Nova Sonic service - format

Browse Source
This commit is contained in:
Paul Kompfner
2025-04-25 15:19:45 -04:00
parent de294caed9
commit 260f7c9b85
1 changed files with 9 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
src/pipecat/services/aws_nova_sonic/aws.py
View File

@@ -61,7 +61,7 @@ class CurrentContent:
type: ContentType type: ContentType
role: Role role: Role
text_stage: TextStage # None if not text text_stage: TextStage # None if not text
text_content: str # starts as None, then fills in if text text_content: str # starts as None, then fills in if text
def __str__(self): def __str__(self):
return ( return (
@@ -388,7 +388,7 @@ class AWSNovaSonicService(LLMService):
type=ContentType(type), type=ContentType(type),
role=Role(role), role=Role(role),
text_stage=TextStage(generation_stage) if generation_stage else None, text_stage=TextStage(generation_stage) if generation_stage else None,
text_content=None text_content=None,
) )
self._content_being_received = content self._content_being_received = content
@@ -396,7 +396,7 @@ class AWSNovaSonicService(LLMService):
if content.type == ContentType.AUDIO: if content.type == ContentType.AUDIO:
# Report that *equivalent* of TTS (this is a speech-to-speech model) started # Report that *equivalent* of TTS (this is a speech-to-speech model) started
# print("[pk] TTS started") # print("[pk] TTS started")
await self.push_frame(TTSStartedFrame()) await self.push_frame(TTSStartedFrame())
print(f"[pk] content start: {self._content_being_received}") print(f"[pk] content start: {self._content_being_received}")
@@ -424,10 +424,7 @@ class AWSNovaSonicService(LLMService):
async def _handle_content_end_event(self, event_json): async def _handle_content_end_event(self, event_json):
content_end = event_json["contentEnd"] content_end = event_json["contentEnd"]
stop_reason = content_end["stopReason"] stop_reason = content_end["stopReason"]
# print( print(f"[pk] content end: {self._content_being_received}.\n stop_reason: {stop_reason}")
# f"[pk] content end: {self._content_being_received}.\n"
# f" stop_reason: {stop_reason}"
# )
# Bookkeeping: clear current content being received # Bookkeeping: clear current content being received
content = self._content_being_received content = self._content_being_received
@@ -443,25 +440,25 @@ class AWSNovaSonicService(LLMService):
# Ignore non-final text, and the "interrupted" message (which isn't meaningful text) # Ignore non-final text, and the "interrupted" message (which isn't meaningful text)
if content.text_stage == TextStage.FINAL and stop_reason != "INTERRUPTED": if content.text_stage == TextStage.FINAL and stop_reason != "INTERRUPTED":
# TODO: the way we're tracking the start and stop of the assistant response here # TODO: the way we're tracking the start and stop of the assistant response here
# is rather busted, and results in way too many "responses" being put into the # is rather busted, and results in way too many "responses" being put into the
# context (every final text content block is treated as its own response). # context (every final text content block is treated as its own response).
# We *should* only record that an assistant response has ended when: # We *should* only record that an assistant response has ended when:
# - the assistant truly finished its turn (stop_reason is END_TURN) # - the assistant truly finished its turn (stop_reason is END_TURN)
# - when this is the next text content block after an INTERRUPTED has occurred # - when this is the next text content block after an INTERRUPTED has occurred
# BUT it seems like there's a bug where, if there are multiple assistant text # BUT it seems like there's a bug where, if there are multiple assistant text
# content blocks, the *first* one gets marked END_TURN rather than the last. # content blocks, the *first* one gets marked END_TURN rather than the last.
print("[pk] LLM full response started") # print("[pk] LLM full response started")
self._assistant_is_responding = True self._assistant_is_responding = True
await self.push_frame(LLMFullResponseStartFrame()) await self.push_frame(LLMFullResponseStartFrame())
if self._assistant_is_responding: if self._assistant_is_responding:
# Add text to the ongoing reported assistant response # Add text to the ongoing reported assistant response
print(f"[pk] LLM text: {content.text_content}") # print(f"[pk] LLM text: {content.text_content}")
await self.push_frame(LLMTextFrame(content.text_content)) await self.push_frame(LLMTextFrame(content.text_content))
# Report that the assistant has finished their response. # Report that the assistant has finished their response.
# TODO: kinda busted. see TODO comment above. # TODO: kinda busted. see TODO comment above.
print("[pk] LLM full response ended") # print("[pk] LLM full response ended")
await self.push_frame(LLMFullResponseEndFrame()) await self.push_frame(LLMFullResponseEndFrame())
self._assistant_is_responding = False self._assistant_is_responding = False