48 lines
1.2 KiB
YAML
48 lines
1.2 KiB
YAML
# Agent behavior configuration for DashScope realtime ASR/TTS.
|
||
# This file only controls agent-side behavior (VAD/LLM/TTS/ASR providers).
|
||
# Infra/server/network settings should stay in .env.
|
||
|
||
agent:
|
||
vad:
|
||
type: silero
|
||
model_path: data/vad/silero_vad.onnx
|
||
threshold: 0.5
|
||
min_speech_duration_ms: 100
|
||
eou_threshold_ms: 800
|
||
|
||
llm:
|
||
# provider: openai | openai_compatible | siliconflow
|
||
provider: openai_compatible
|
||
model: deepseek-v3
|
||
temperature: 0.7
|
||
api_key: your_llm_api_key
|
||
api_url: https://api.qnaigc.com/v1
|
||
|
||
tts:
|
||
provider: dashscope
|
||
api_key: your_tts_api_key
|
||
api_url: wss://dashscope.aliyuncs.com/api-ws/v1/realtime
|
||
model: qwen3-tts-flash-realtime
|
||
voice: Cherry
|
||
dashscope_mode: commit
|
||
speed: 1.0
|
||
|
||
asr:
|
||
provider: dashscope
|
||
api_key: your_asr_api_key
|
||
api_url: wss://dashscope.aliyuncs.com/api-ws/v1/realtime
|
||
model: qwen3-asr-flash-realtime
|
||
interim_interval_ms: 500
|
||
min_audio_ms: 300
|
||
start_min_speech_ms: 160
|
||
pre_speech_ms: 240
|
||
final_tail_ms: 120
|
||
|
||
duplex:
|
||
enabled: true
|
||
system_prompt: 你是一个人工智能助手，你用简答语句回答，避免使用标点符号和emoji。
|
||
|
||
barge_in:
|
||
min_duration_ms: 200
|
||
silence_tolerance_ms: 60
|