feat(qwen-audio): add Alibaba Cloud Qwen-Audio Realtime service integration
- Introduced Qwen-Audio Realtime service for speech-to-speech processing in Pipecat. - Updated interface catalog to include Qwen-Audio Realtime capabilities. - Enhanced model resource testing to support new service. - Added configuration options for audio sample rates and turn detection modes. - Updated documentation to reflect integration details and usage instructions.
This commit is contained in:
@@ -119,6 +119,58 @@ INTERFACE_DEFINITIONS: list[dict] = [
|
||||
),
|
||||
],
|
||||
},
|
||||
{
|
||||
"interface_type": "qwen-audio-realtime",
|
||||
"name": "Alibaba Cloud Qwen-Audio Realtime",
|
||||
"capability": "Realtime",
|
||||
"fields": [
|
||||
field(
|
||||
"modelId",
|
||||
"Model ID",
|
||||
type_="select",
|
||||
required=True,
|
||||
default="qwen-audio-3.0-realtime-flash",
|
||||
options=[
|
||||
"qwen-audio-3.0-realtime-flash",
|
||||
"qwen-audio-3.0-realtime-plus",
|
||||
],
|
||||
),
|
||||
field(
|
||||
"apiUrl",
|
||||
"WebSocket URL (with Workspace ID)",
|
||||
type_="url",
|
||||
required=True,
|
||||
),
|
||||
field(
|
||||
"apiKey",
|
||||
"API Key",
|
||||
group="secrets",
|
||||
type_="password",
|
||||
required=True,
|
||||
),
|
||||
field("voice", "Voice", default="longanqian"),
|
||||
field(
|
||||
"turnDetection",
|
||||
"Turn Detection",
|
||||
type_="select",
|
||||
default="server_vad",
|
||||
options=["server_vad", "smart_turn"],
|
||||
),
|
||||
field("vadThreshold", "VAD Threshold", type_="number", default=0.5),
|
||||
field(
|
||||
"silenceDurationMs",
|
||||
"VAD Silence Duration (ms)",
|
||||
type_="number",
|
||||
default=800,
|
||||
),
|
||||
field(
|
||||
"maxHistoryTurns",
|
||||
"Max History Turns",
|
||||
type_="number",
|
||||
default=20,
|
||||
),
|
||||
],
|
||||
},
|
||||
{
|
||||
"interface_type": "xfyun-asr",
|
||||
"name": "Xfyun Streaming ASR",
|
||||
|
||||
Reference in New Issue
Block a user