360 Secure SkillHub
360 安全技能中心

Aliyun Qwen Tts Voice Clone

这是阿里云Model Studio的Qwen TTS语音克隆工具,核心是通过指定模型从低噪干净的音频样本克隆语音音色,生成可复用的voice_id,支持合成指定文本的语音,也可选择流式输出。典型场景包括个性化有声书制作、智能客服语音定制等。输入需提供文本和语音样本,可选语音名称、流式参数;输出含音频链接/流式PCM片段、voice_id及请求ID。使用前需安装SDK并配置API密钥,需遵守合规要求。

cinience
cinience
热度 34.93
最近更新 2026-05-11 17:39:08
分类:创作依赖:代码执行 / 第三方 API / 读写文件

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add aliyun-qwen-tts-voice-clone
Metadata
name
aliyun-qwen-tts-voice-clone
description
Use when cloning voices with Alibaba Cloud Model Studio Qwen TTS VC models. Use when creating cloned voices from sample audio and synthesizing text with cloned timbre.
version
1.0.0

Category: provider

Model Studio Qwen TTS Voice Clone

Use voice cloning models to replicate timbre from enrollment audio samples.

Critical model names

Use one of these exact model strings:

  • qwen3-tts-vc-2026-01-22
  • qwen3-tts-vc-realtime-2026-01-15

Prerequisites

  • Install SDK in a virtual environment:
python3 -m venv .venv
. .venv/bin/activate
python -m pip install dashscope
  • Set DASHSCOPE_API_KEY in your environment, or add dashscope_api_key to ~/.alibabacloud/credentials.

Normalized interface (tts.voice_clone)

Request

  • text (string, required)
  • voice_sample (string | bytes, required) enrollment sample
  • voice_name (string, optional)
  • stream (bool, optional)

Response

  • audio_url (string) or streaming PCM chunks
  • voice_id (string)
  • request_id (string)

Operational guidance

  • Use clean speech samples with low background noise.
  • Respect consent and policy requirements for cloned voices.
  • Persist generated voice_id and reuse for future synthesis requests.

Local helper script

Prepare a normalized request JSON and validate response schema:

.venv/bin/python skills/ai/audio/aliyun-qwen-tts-voice-clone/scripts/prepare_voice_clone_request.py \
  --text "Welcome to this voice-clone demo" \
  --voice-sample "https://example.com/voice-sample.wav"

Output location

  • Default output: output/ai-audio-tts-voice-clone/audio/
  • Override base dir with OUTPUT_DIR.

Validation

mkdir -p output/aliyun-qwen-tts-voice-clone
for f in skills/ai/audio/aliyun-qwen-tts-voice-clone/scripts/*.py; do
  python3 -m py_compile "$f"
done
echo "py_compile_ok" > output/aliyun-qwen-tts-voice-clone/validate.txt

Pass criteria: command exits 0 and output/aliyun-qwen-tts-voice-clone/validate.txt is generated.

Output And Evidence

  • Save artifacts, command outputs, and API response summaries under output/aliyun-qwen-tts-voice-clone/.
  • Include key parameters (region/resource id/time range) in evidence files for reproducibility.

Workflow

  1. Confirm user intent, region, identifiers, and whether the operation is read-only or mutating.
  2. Run one minimal read-only query first to verify connectivity and permissions.
  3. Execute the target operation with explicit parameters and bounded scope.
  4. Verify results and save output/evidence files.

References

  • references/sources.md