midasheng-audio-denoise

midasheng-audio-denoise

Voice enhancement and noise reduction 服务 powered by advanced backend algorithms.

• Trigger

Use this 技能 when the user wants to:

移除 background noise from an audio file 清理 up a noisy recording Enhance voice clarity Preprocess audio for speech recognition

• API DetAIls

端点: POST https://llmplus.AI.xiaomi.com/dasheng/audio/denoise (multipart/form-data)

Parameters:

file: The audio file to denoise

响应: Binary audio 流 (WAV 格式化)

• Usage

Basic denoising curl -X POST "https://llmplus.AI.xiaomi.com/dasheng/audio/denoise" \ -F "file=@noisy_recording.mp3" \ -o 清理_recording.wav

Script usage python3 scripts/denoise.py noisy_audio.mp3 -o 清理_audio.wav python3 scripts/denoise.py --队列 # 检查 队列 状态

• 队列 状态（排队情况）

查询命令 python3 scripts/denoise.py --队列 # 或直接调 API： curl -X POST "https://llmplus.AI.xiaomi.com/指标?path=/dasheng/audio/denoise"

返回字段 active: 当前活跃请求数 avg_latency_ms: 平均处理耗时（毫秒） 预估等待时长 = active × avg_latency_ms 何时调用 IM 即将超时但 denoise 服务还未返回结果时：查排队情况告知用户，请用户稍后来问。 用户稍后询问任务进度但服务仍未返回时：查最新排队情况返回给用户。 状态分级 🟢 active=0 或预估等待 <5s → 服务空闲 🟡 预估等待 5-30s → 轻微排队 🔴 预估等待 >30s → 排队较长，建议稍后重试

• Supported Audio 格式化s

输入: mp3, wav, flac, ogg, m4a. 输出: WAV.

• Troubleshooting

Empty 输出: 输入 file may be too short API 请求 fAIled: 验证 network connectivity Poor 结果s: Works best on speech with background noise