feat: add recording audio option in tool and node transitions (#232)

* feat: allow uploading recording as part of node transition

* feat: allow recordings in tool transitions

* chore: fix tests
This commit is contained in:
Abhishek 2026-04-10 17:53:42 +05:30 committed by GitHub
parent 3f19a16e7f
commit 7c245051d2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
54 changed files with 3575 additions and 640 deletions

View file

@ -178,6 +178,11 @@ async def initiate_call(
workflow_run_id = request.workflow_run_id
if not workflow_run_id:
# Fetch workflow to merge template context variables (e.g. caller_number,
# called_number set in workflow settings for testing pre-call data fetch)
workflow = await db_client.get_workflow_by_id(request.workflow_id)
template_vars = (workflow.template_context_variables or {}) if workflow else {}
numeric_suffix = int(str(uuid.uuid4()).replace("-", "")[:8], 16) % 100000000
workflow_run_name = f"WR-TEL-OUT-{numeric_suffix:08d}"
workflow_run = await db_client.create_workflow_run(
@ -187,6 +192,7 @@ async def initiate_call(
user_id=user.id,
call_type=CallType.OUTBOUND,
initial_context={
**template_vars,
"phone_number": phone_number,
"called_number": phone_number,
"provider": provider.PROVIDER_NAME,

View file

@ -45,17 +45,29 @@ class HttpApiConfig(BaseModel):
timeout_ms: Optional[int] = Field(
default=5000, description="Request timeout in milliseconds"
)
customMessage: Optional[str] = Field(
default=None, description="Custom message to play after tool execution"
)
customMessageType: Optional[Literal["text", "audio"]] = Field(
default=None, description="Type of custom message: text or audio"
)
customMessageRecordingId: Optional[str] = Field(
default=None, description="Recording ID for audio custom message"
)
class EndCallConfig(BaseModel):
"""Configuration for End Call tools."""
messageType: Literal["none", "custom"] = Field(
messageType: Literal["none", "custom", "audio"] = Field(
default="none", description="Type of goodbye message"
)
customMessage: Optional[str] = Field(
default=None, description="Custom message to play before ending the call"
)
audioRecordingId: Optional[str] = Field(
default=None, description="Recording ID for audio goodbye message"
)
endCallReason: bool = Field(
default=False,
description="When enabled, LLM must provide a reason for ending the call. "
@ -74,12 +86,15 @@ class TransferCallConfig(BaseModel):
destination: str = Field(
description="Phone number or SIP endpoint to transfer the call to (E.164 format e.g., +1234567890, or SIP endpoint e.g., PJSIP/1234)"
)
messageType: Literal["none", "custom"] = Field(
messageType: Literal["none", "custom", "audio"] = Field(
default="none", description="Type of message to play before transfer"
)
customMessage: Optional[str] = Field(
default=None, description="Custom message to play before transferring the call"
)
audioRecordingId: Optional[str] = Field(
default=None, description="Recording ID for audio message before transfer"
)
timeout: int = Field(
default=30,
ge=5,

View file

@ -16,6 +16,7 @@ from api.schemas.workflow_recording import (
BatchRecordingUploadResponseSchema,
RecordingListResponseSchema,
RecordingResponseSchema,
RecordingUpdateRequestSchema,
RecordingUploadResponseSchema,
)
from api.services.auth.depends import get_user
@ -25,11 +26,11 @@ from api.services.storage import storage_fs
router = APIRouter(prefix="/workflow-recordings", tags=["workflow-recordings"])
async def _generate_unique_recording_id() -> str:
"""Generate a globally unique short recording ID."""
async def _generate_unique_recording_id(organization_id: int) -> str:
"""Generate a unique short recording ID within an organization."""
for _ in range(10):
rid = generate_short_id(8)
exists = await db_client.check_recording_id_exists(rid)
exists = await db_client.check_recording_id_exists(rid, organization_id)
if not exists:
return rid
raise HTTPException(
@ -69,11 +70,13 @@ async def get_upload_urls(
try:
items = []
for fd in request.files:
recording_id = await _generate_unique_recording_id()
recording_id = await _generate_unique_recording_id(
user.selected_organization_id
)
storage_key = (
f"recordings/{user.selected_organization_id}"
f"/{request.workflow_id}/{recording_id}"
f"/{recording_id}"
f"/{fd.filename}"
)
@ -100,7 +103,7 @@ async def get_upload_urls(
logger.info(
f"Generated {len(items)} recording upload URL(s), "
f"workflow {request.workflow_id}, org {user.selected_organization_id}"
f"org {user.selected_organization_id}"
)
return BatchRecordingUploadResponseSchema(items=items)
@ -131,22 +134,20 @@ async def create_recordings(
for rec_req in request.recordings:
recording = await db_client.create_recording(
recording_id=rec_req.recording_id,
workflow_id=rec_req.workflow_id,
organization_id=user.selected_organization_id,
tts_provider=rec_req.tts_provider,
tts_model=rec_req.tts_model,
tts_voice_id=rec_req.tts_voice_id,
transcript=rec_req.transcript,
storage_key=rec_req.storage_key,
storage_backend=backend.value,
created_by=user.id,
tts_provider=rec_req.tts_provider,
tts_model=rec_req.tts_model,
tts_voice_id=rec_req.tts_voice_id,
metadata=rec_req.metadata,
)
results.append(_build_response(recording))
logger.info(
f"Created {len(results)} recording(s) for "
f"workflow {request.recordings[0].workflow_id}"
f"Created {len(results)} recording(s) for org {user.selected_organization_id}"
)
return BatchRecordingCreateResponseSchema(recordings=results)
@ -163,10 +164,12 @@ async def create_recordings(
@router.get(
"/",
response_model=RecordingListResponseSchema,
summary="List recordings for a workflow",
summary="List recordings",
)
async def list_recordings(
workflow_id: Annotated[int, Query(description="Workflow ID")],
workflow_id: Annotated[
Optional[int], Query(description="Filter by workflow ID")
] = None,
tts_provider: Annotated[
Optional[str], Query(description="Filter by TTS provider")
] = None,
@ -178,11 +181,11 @@ async def list_recordings(
] = None,
user=Depends(get_user),
):
"""List recordings for a workflow, optionally filtered by TTS configuration."""
"""List recordings for the organization, optionally filtered."""
try:
recordings = await db_client.get_recordings_for_workflow(
workflow_id=workflow_id,
recordings = await db_client.get_recordings(
organization_id=user.selected_organization_id,
workflow_id=workflow_id,
tts_provider=tts_provider,
tts_model=tts_model,
tts_voice_id=tts_voice_id,
@ -233,6 +236,74 @@ async def delete_recording(
) from exc
@router.patch(
"/{id}",
response_model=RecordingResponseSchema,
summary="Update a recording's Recording ID",
)
async def update_recording(
id: int,
request: RecordingUpdateRequestSchema,
user=Depends(get_user),
):
"""Update the recording_id (descriptive name) of a recording."""
try:
new_id = request.recording_id.strip()
if not new_id:
raise HTTPException(status_code=400, detail="Recording ID cannot be empty")
existing = await db_client.get_recording_by_id(
id, user.selected_organization_id
)
if not existing:
raise HTTPException(status_code=404, detail="Recording not found")
if new_id == existing.recording_id:
return _build_response(existing)
exists = await db_client.check_recording_id_exists(
new_id, user.selected_organization_id
)
if exists:
raise HTTPException(
status_code=409,
detail=f"Recording ID '{new_id}' is already in use",
)
old_id = existing.recording_id
recording = await db_client.update_recording_id(
id=id,
new_recording_id=new_id,
organization_id=user.selected_organization_id,
)
if not recording:
raise HTTPException(status_code=404, detail="Recording not found")
# Replace old recording ID in all non-legacy workflow definitions
updated = await db_client.replace_recording_id_in_workflows(
old_id=old_id,
new_id=new_id,
organization_id=user.selected_organization_id,
)
if updated:
logger.info(
f"Updated {updated} workflow definition(s) with new recording ID "
f"'{old_id}' -> '{new_id}'"
)
return _build_response(recording)
except HTTPException:
raise
except Exception as exc:
logger.error(f"Error updating recording: {exc}")
raise HTTPException(
status_code=500, detail="Failed to update recording"
) from exc
@router.post(
"/transcribe",
summary="Transcribe an audio file",