|
503 | 503 | ] |
504 | 504 | } |
505 | 505 | ] |
506 | | - }, |
| 506 | + }, |
507 | 507 | { |
508 | 508 | "name": "voice_assistant_realtime", |
509 | 509 | "auto_start": true, |
|
1344 | 1344 | ] |
1345 | 1345 | } |
1346 | 1346 | ] |
| 1347 | + }, |
| 1348 | + { |
| 1349 | + "name": "voice_assistant_multimodal", |
| 1350 | + "auto_start": true, |
| 1351 | + "nodes": [ |
| 1352 | + { |
| 1353 | + "type": "extension", |
| 1354 | + "name": "agora_rtc", |
| 1355 | + "addon": "agora_rtc", |
| 1356 | + "extension_group": "default", |
| 1357 | + "property": { |
| 1358 | + "app_id": "${env:AGORA_APP_ID}", |
| 1359 | + "token": "<agora_token>", |
| 1360 | + "channel": "ten_agent_test", |
| 1361 | + "stream_id": 1234, |
| 1362 | + "remote_stream_id": 123, |
| 1363 | + "subscribe_audio": true, |
| 1364 | + "publish_audio": true, |
| 1365 | + "publish_data": true, |
| 1366 | + "enable_agora_asr": false, |
| 1367 | + "agora_asr_vendor_name": "microsoft", |
| 1368 | + "agora_asr_language": "en-US", |
| 1369 | + "agora_asr_vendor_key": "${env:AZURE_STT_KEY|}", |
| 1370 | + "agora_asr_vendor_region": "${env:AZURE_STT_REGION|}", |
| 1371 | + "agora_asr_session_control_file_path": "session_control.conf", |
| 1372 | + "subscribe_video_pix_fmt": 4, |
| 1373 | + "subscribe_video": true, |
| 1374 | + "max_memory_length": 10 |
| 1375 | + } |
| 1376 | + }, |
| 1377 | + { |
| 1378 | + "type": "extension", |
| 1379 | + "name": "stt", |
| 1380 | + "addon": "transcribe_asr_python", |
| 1381 | + "extension_group": "stt", |
| 1382 | + "property": { |
| 1383 | + "access_key": "${env:AWS_ACCESS_KEY_ID}", |
| 1384 | + "lang_code": "en-US", |
| 1385 | + "region": "us-east-1", |
| 1386 | + "sample_rate": "16000", |
| 1387 | + "secret_key": "${env:AWS_SECRET_ACCESS_KEY}" |
| 1388 | + } |
| 1389 | + }, |
| 1390 | + { |
| 1391 | + "type": "extension", |
| 1392 | + "name": "llm", |
| 1393 | + "addon": "bedrock_llm_python", |
| 1394 | + "extension_group": "chatgpt", |
| 1395 | + "property": { |
| 1396 | + "access_key_id": "${env:AWS_ACCESS_KEY_ID}", |
| 1397 | + "greeting": "TEN Agent connected. I am nova, How can I help you today?", |
| 1398 | + "max_memory_length": 10, |
| 1399 | + "max_tokens": 256, |
| 1400 | + "model": "us.amazon.nova-lite-v1:0", |
| 1401 | + "prompt": "Now you are an intelligent assistant with real-time interaction capabilities. I will provide you with a series of real-time video image information. Please understand these images as video frames. Based on the images and the user's input, engage in a conversation with the user, remembering the dialogue content in a concise and clear manner.", |
| 1402 | + "region": "us-east-1", |
| 1403 | + "secret_access_key": "${env:AWS_SECRET_ACCESS_KEY}", |
| 1404 | + "temperature": 0.7, |
| 1405 | + "topK": 10, |
| 1406 | + "topP": 0.5, |
| 1407 | + "is_memory_enabled": false, |
| 1408 | + "is_enable_video": true |
| 1409 | + } |
| 1410 | + }, |
| 1411 | + { |
| 1412 | + "type": "extension", |
| 1413 | + "name": "tts", |
| 1414 | + "addon": "polly_tts", |
| 1415 | + "extension_group": "tts", |
| 1416 | + "property": { |
| 1417 | + "region": "us-east-1", |
| 1418 | + "access_key": "${env:AWS_ACCESS_KEY_ID}", |
| 1419 | + "secret_key": "${env:AWS_SECRET_ACCESS_KEY}", |
| 1420 | + "engine": "generative", |
| 1421 | + "voice": "Ruth", |
| 1422 | + "sample_rate": 16000, |
| 1423 | + "lang_code": "en-US" |
| 1424 | + } |
| 1425 | + }, |
| 1426 | + { |
| 1427 | + "type": "extension", |
| 1428 | + "name": "interrupt_detector", |
| 1429 | + "addon": "interrupt_detector_python", |
| 1430 | + "extension_group": "default", |
| 1431 | + "property": {} |
| 1432 | + }, |
| 1433 | + { |
| 1434 | + "type": "extension", |
| 1435 | + "name": "message_collector", |
| 1436 | + "addon": "message_collector", |
| 1437 | + "extension_group": "transcriber", |
| 1438 | + "property": {} |
| 1439 | + } |
| 1440 | + ], |
| 1441 | + "connections": [ |
| 1442 | + { |
| 1443 | + "extension": "agora_rtc", |
| 1444 | + "cmd": [ |
| 1445 | + { |
| 1446 | + "name": "on_user_joined", |
| 1447 | + "dest": [ |
| 1448 | + { |
| 1449 | + "extension": "llm" |
| 1450 | + } |
| 1451 | + ] |
| 1452 | + }, |
| 1453 | + { |
| 1454 | + "name": "on_user_left", |
| 1455 | + "dest": [ |
| 1456 | + { |
| 1457 | + "extension": "llm" |
| 1458 | + } |
| 1459 | + ] |
| 1460 | + }, |
| 1461 | + { |
| 1462 | + "name": "on_connection_failure", |
| 1463 | + "dest": [ |
| 1464 | + { |
| 1465 | + "extension": "llm" |
| 1466 | + } |
| 1467 | + ] |
| 1468 | + } |
| 1469 | + ], |
| 1470 | + "audio_frame": [ |
| 1471 | + { |
| 1472 | + "name": "pcm_frame", |
| 1473 | + "dest": [ |
| 1474 | + { |
| 1475 | + "extension": "stt" |
| 1476 | + } |
| 1477 | + ] |
| 1478 | + } |
| 1479 | + ], |
| 1480 | + "video_frame": [ |
| 1481 | + { |
| 1482 | + "name": "video_frame", |
| 1483 | + "dest": [ |
| 1484 | + { |
| 1485 | + "extension": "llm" |
| 1486 | + } |
| 1487 | + ] |
| 1488 | + } |
| 1489 | + ] |
| 1490 | + }, |
| 1491 | + { |
| 1492 | + "extension": "stt", |
| 1493 | + "data": [ |
| 1494 | + { |
| 1495 | + "name": "text_data", |
| 1496 | + "dest": [ |
| 1497 | + { |
| 1498 | + "extension": "interrupt_detector" |
| 1499 | + }, |
| 1500 | + { |
| 1501 | + "extension": "message_collector" |
| 1502 | + } |
| 1503 | + ] |
| 1504 | + } |
| 1505 | + ] |
| 1506 | + }, |
| 1507 | + { |
| 1508 | + "extension": "llm", |
| 1509 | + "cmd": [ |
| 1510 | + { |
| 1511 | + "name": "flush", |
| 1512 | + "dest": [ |
| 1513 | + { |
| 1514 | + "extension": "tts" |
| 1515 | + } |
| 1516 | + ] |
| 1517 | + } |
| 1518 | + ], |
| 1519 | + "data": [ |
| 1520 | + { |
| 1521 | + "name": "text_data", |
| 1522 | + "dest": [ |
| 1523 | + { |
| 1524 | + "extension": "tts" |
| 1525 | + }, |
| 1526 | + { |
| 1527 | + "extension": "message_collector" |
| 1528 | + } |
| 1529 | + ] |
| 1530 | + } |
| 1531 | + ] |
| 1532 | + }, |
| 1533 | + { |
| 1534 | + "extension": "message_collector", |
| 1535 | + "data": [ |
| 1536 | + { |
| 1537 | + "name": "data", |
| 1538 | + "dest": [ |
| 1539 | + { |
| 1540 | + "extension": "agora_rtc" |
| 1541 | + } |
| 1542 | + ] |
| 1543 | + } |
| 1544 | + ] |
| 1545 | + }, |
| 1546 | + { |
| 1547 | + "extension": "tts", |
| 1548 | + "cmd": [ |
| 1549 | + { |
| 1550 | + "name": "flush", |
| 1551 | + "dest": [ |
| 1552 | + { |
| 1553 | + "extension": "agora_rtc" |
| 1554 | + } |
| 1555 | + ] |
| 1556 | + } |
| 1557 | + ], |
| 1558 | + "audio_frame": [ |
| 1559 | + { |
| 1560 | + "name": "pcm_frame", |
| 1561 | + "dest": [ |
| 1562 | + { |
| 1563 | + "extension": "agora_rtc" |
| 1564 | + } |
| 1565 | + ] |
| 1566 | + } |
| 1567 | + ] |
| 1568 | + }, |
| 1569 | + { |
| 1570 | + "extension": "interrupt_detector", |
| 1571 | + "cmd": [ |
| 1572 | + { |
| 1573 | + "name": "flush", |
| 1574 | + "dest": [ |
| 1575 | + { |
| 1576 | + "extension": "llm" |
| 1577 | + } |
| 1578 | + ] |
| 1579 | + } |
| 1580 | + ], |
| 1581 | + "data": [ |
| 1582 | + { |
| 1583 | + "name": "text_data", |
| 1584 | + "dest": [ |
| 1585 | + { |
| 1586 | + "extension": "llm" |
| 1587 | + } |
| 1588 | + ] |
| 1589 | + } |
| 1590 | + ] |
| 1591 | + } |
| 1592 | + ] |
1347 | 1593 | } |
1348 | 1594 | ], |
1349 | 1595 | "log_level": 3 |
|
0 commit comments