agentscope.realtimeParameters
class Parameters(RealtimeModelBase.Parameters)agentscope/realtime/_openai/_model.py:28–61参数，处理低延迟的实时语音与多模态交互。位于 agentscope.realtime 包，继承自 RealtimeModelBase.Parameters。
Tuneables surfaced to the UI. Declared in agentscope.realtime. Extends RealtimeModelBase.Parameters.
属性摘要Attribute Summary
| 属性Attribute | 类型Type | 说明Description | 值Value |
|---|---|---|---|
voice | Literal['alloy', 'ash', 'ballad', 'cedar', 'coral', 'echo', 'marin', 'sage', 'shimmer', 'verse'] | 语音字段。 | Field(default='marin', title='Voice') |
turn_detection | Literal['server_vad', 'semantic_vad', 'none'] | `turn_detection` 字段。 | Field(default='server_vad', title='Turn Detection', description='``none`` hands endpointing to the caller.') |
vad_threshold | float | 语音活动检测阈值字段。 | Field(default=0.5, ge=0.0, le=1.0) |
vad_prefix_padding_ms | int | `vad_prefix_padding_ms` 字段。 | Field(default=300, ge=0) |
vad_silence_duration_ms | int | `vad_silence_duration_ms` 字段。 | Field(default=500, ge=0) |
vad_eagerness | Literal['low', 'medium', 'high', 'auto'] | `vad_eagerness` 字段。 | Field(default='auto', title='Semantic VAD Eagerness', description='How soon ``semantic_vad`` decides the user is done.') |
input_audio_transcription | str | `input_audio_transcription` 字段。 | Field(default='gpt-4o-mini-transcribe', title='Input Transcription Model', description="The ASR model transcribing the user's speech; an empty string turns transcription off.") |
属性详情Attribute Detail
voice
Literal['alloy', 'ash', 'ballad', 'cedar', 'coral', 'echo', 'marin', 'sage', 'shimmer', 'verse'] = Field(default='marin', title='Voice')语音字段。声明为 Literal['alloy', 'ash', 'ballad', 'cedar', 'coral', 'echo', 'marin', 'sage', 'shimmer', 'verse']。
Voice
agentscope/realtime/_openai/_model.py:31turn_detection
Literal['server_vad', 'semantic_vad', 'none'] = Field(default='server_vad', title='Turn Detection', description='``none`` hands endpointing to the caller.')turn_detection 字段。声明为 Literal['server_vad', 'semantic_vad', 'none']。
none hands endpointing to the caller.
agentscope/realtime/_openai/_model.py:43vad_threshold
float = Field(default=0.5, ge=0.0, le=1.0)语音活动检测阈值字段。声明为 float。
The vad threshold field.
agentscope/realtime/_openai/_model.py:48vad_prefix_padding_ms
int = Field(default=300, ge=0)vad_prefix_padding_ms 字段。声明为 int。
The vad prefix padding ms field.
agentscope/realtime/_openai/_model.py:49vad_silence_duration_ms
int = Field(default=500, ge=0)vad_silence_duration_ms 字段。声明为 int。
The vad silence duration ms field.
agentscope/realtime/_openai/_model.py:50vad_eagerness
Literal['low', 'medium', 'high', 'auto'] = Field(default='auto', title='Semantic VAD Eagerness', description='How soon ``semantic_vad`` decides the user is done.')vad_eagerness 字段。声明为 Literal['low', 'medium', 'high', 'auto']。
How soon semantic_vad decides the user is done.
agentscope/realtime/_openai/_model.py:51input_audio_transcription
str = Field(default='gpt-4o-mini-transcribe', title='Input Transcription Model', description="The ASR model transcribing the user's speech; an empty string turns transcription off.")input_audio_transcription 字段。声明为 str。
The ASR model transcribing the user's speech; an empty string turns transcription off.
agentscope/realtime/_openai/_model.py:56