Files
AI-VideoAssistant/docs/content/customization/asr.md
Xin Wang aeeeee20d1 Add Volcengine support for TTS and ASR services
- Introduced Volcengine as a new provider for both TTS and ASR services.
- Updated configuration files to include Volcengine-specific parameters such as app_id, resource_id, and uid.
- Enhanced the ASR service to support streaming mode with Volcengine's API.
- Modified existing tests to validate the integration of Volcengine services.
- Updated documentation to reflect the addition of Volcengine as a supported provider for TTS and ASR.
- Refactored service factory to accommodate Volcengine alongside existing providers.
2026-03-08 23:09:50 +08:00

1.3 KiB
Raw Blame History

语音识别

语音识别ASR负责将用户音频实时转写为文本供对话引擎理解。

模式

  • offline:引擎本地缓冲音频后触发识别(适用于 OpenAI-compatible / SiliconFlow
  • streaming:音频分片实时发送到服务端,服务端持续返回转写事件(适用于 DashScope Realtime ASR、Volcengine BigASR

配置项

配置项 说明
ASR 引擎 选择语音识别服务提供商
模型 识别模型名称
enable_interim 是否开启离线 ASR 中间结果(默认 false,仅离线模式生效)
app_id / resource_id Volcengine 等厂商的应用标识与资源标识
request_params 厂商原生请求参数透传,例如 end_window_sizeforce_to_speech_timecontext
语言 中文/英文/多语言
热词 提升特定词汇识别准确率
标点与规范化 是否自动补全标点、文本规范化

建议

  • 客服场景建议开启热词并维护业务词表
  • 多语言场景建议按会话入口显式指定语言
  • 对延迟敏感场景优先选择流式识别模型
  • 当前支持提供商:openai_compatiblesiliconflowdashscopevolcenginebuffered(回退)

相关文档