IMA Studio Music Generation

IMA Voice AI Creation

Scope & Dependencies (Declared for Transparency)

• Credentials: This skill requires an IMA API key at runtime (IMA_API_KEY or --api-key). The key is sent only to api.imastudio.com. Obtain keys at https://imastudio.com. Declared in registry as required.
• Optional dependency: When ima-knowledge-ai is installed, this skill may instruct the agent to read that skill's reference files (~/.openclaw/skills/ima-knowledge-ai/references/*) for workflow and model-selection guidance. This skill is self-contained — it works fully without ima-knowledge-ai. Reading another skill's files is optional and only for complex or multi-step tasks; users who do not have or trust ima-knowledge-ai can ignore those steps and use this skill's built-in defaults and 📥 User Input Parsing tables.
• Local paths: This skill reads/writes ~/.openclaw/memory/ima_prefs.json (preferences) and ~/.openclaw/logs/ima_skills/ (logs; auto-deleted after 7 days). User can delete these anytime.

Optional: Read Knowledge Base (When ima-knowledge-ai Is Installed)

If ima-knowledge-ai is not installed: Skip this section. Use only this SKILL's default models and the 📥 User Input Parsing tables for model_id and parameters.

When ima-knowledge-ai is installed and the task is complex, you may optionally read its reference files for better workflow and model choice:

1. Workflow complexity — Read ima-knowledge-ai/references/workflow-design.md if:

   • User mentions: "MV"、"配乐"、"完整作品"、"多步骤"、"soundtrack"
   • Task involves: video + music coordination, multi-track production, integrated workflows
   • Complex requirements that need task decomposition

2. Model selection — Read ima-knowledge-ai/references/model-selection.md if:

   • Unsure which model to use (Suno vs DouBao BGM vs DouBao Song)
   • Need cost/quality trade-off guidance
   • User specifies budget or quality requirements

Why this is optional:

• Music generation is often part of a larger workflow (video + music, story + soundtrack)
• For simple single-track requests, proceed directly with this skill's defaults
• For complex workflows, reading the knowledge base can improve task decomposition and model choice

Example workflow case (when using optional knowledge base):

User: "帮我做个产品宣传MV，有背景音乐"

❌ Wrong: 直接生成音乐 (music alone, no coordination with video)

✅ Right (if ima-knowledge-ai available): 
  1. Read workflow-design.md
  2. Decompose: Script → Video shots → Background music (matching video duration/mood)
  3. Generate video first (get duration)
  4. Generate BGM with matching duration and style

How to check (optional):

# Only if ima-knowledge-ai is installed and task is complex
if ima_knowledge_ai_installed and (complex_workflow or multi_step):
    read("~/.openclaw/skills/ima-knowledge-ai/references/workflow-design.md")

if ima_knowledge_ai_installed and unsure_model_choice:
    read("~/.openclaw/skills/ima-knowledge-ai/references/model-selection.md")

# Choose model (this skill's logic works with or without knowledge base)
if "background music" or "BGM" or "instrumental":
    use_doubao_bgm()  # 30pts, pure instrumental
elif "song" or "lyrics" or "vocals":
    use_suno_sonic()  # 25pts, full-featured with lyrics
else:
    use_suno_sonic()  # Default: most versatile

For simple requests: Proceed directly with this skill's defaults. No need to read other skills' files.

📥 User Input Parsing (Model & Parameter Recognition)

Purpose: So that any agent (Claude or other models) parses user intent consistently, follow these rules when deriving model_id and task type from natural language. Normalize first, then map.

1. User phrasing → model selection (model_id)

If the user does not specify, default to Suno (sonic) for versatility. For "背景音乐"/"BGM"/"配乐" only → use DouBao BGM (GenBGM).

2. Music-specific parameters (Suno)

When using Suno with lyrics or vocal control, set custom_mode: true and pass lyrics / vocal_gender per API docs.

⚙️ How This Skill Works

For transparency: This skill uses a bundled Python script (scripts/ima_voice_create.py) to call the IMA Open API. The script:

• Sends your prompt to https://api.imastudio.com (IMA's servers)
• Uses --user-id only locally as a key for storing your model preferences
• Returns a music URL when generation is complete
• NEW (v1.1.0): Automatic reflection mechanism — if generation fails, the script automatically retries up to 3 times with smart parameter adjustments

🧠 Reflection Mechanism (Automatic Error Recovery)

This skill now includes an intelligent reflection system that automatically recovers from common errors:

3-Layer Retry Strategy:

1. Attempt 1: Original Parameters

   • Uses your provided parameters with smart credit_rule selection
   • Most tasks succeed on first try

2. Attempt 2: Strict Match (Error 6009 Fix)

   • Automatically removes unsupported parameters
   • Only keeps parameters in credit_rules.attributes
   • Example: Removes unsupported Suno parameters if not in model config

3. Attempt 3: Fallback to Default (Error 6010 Fix)

   • Uses model's default configuration
   • Uses credit_rules[0] (first rule = safest default)
   • Guarantees maximum compatibility

Common Errors Fixed Automatically:

• Error 6009: "No exact rule match found for parameters" → removes unsupported params
• Error 6010: "Attribute ID does not match" → corrects attribute_id to match params
• Invalid product attribute: Uses default rule configuration

User Experience:

• ✅ Transparent: Shows reflection log when retries happen
• ✅ Fast: Most tasks succeed on first attempt (no delay)
• ✅ Smart: Learns from errors and adjusts automatically
• ✅ User-friendly: Provides helpful suggestions if all 3 attempts fail

Example Output (with reflection):

🚀 Creating music generation task…

🧠 反省日志 (2 次尝试):
   ❌ [尝试 1] Invalid product attribute → 移除不支持的参数: ['unsupported_param']
   ✅ [尝试 2] ✅ 成功（尝试 2）

✅ Task created: task_abc123xyz

What gets sent to IMA servers:

• ✅ Your music prompt/description
• ✅ Model selection (Suno/DouBao)
• ❌ NO API key in prompts (key is used for authentication only)
• ❌ NO user_id (it's only used locally)

What's stored locally:

• ~/.openclaw/memory/ima_prefs.json - Your model preferences (< 1 KB)
• ~/.openclaw/logs/ima_skills/ - Generation logs (auto-deleted after 7 days)

Agent Execution (Internal Reference)

Note for users: You can review the script source at scripts/ima_voice_create.py anytime.
This skill uses only api.imastudio.com (no image upload needed for music generation, unlike image/video skills which also use imapi.liveme.com).

Use the bundled script internally:

# Generate music — Suno sonic-v5
python3 {baseDir}/scripts/ima_voice_create.py \
  --api-key  $IMA_API_KEY \
  --task-type text_to_music \
  --model-id  sonic \
  --prompt   "upbeat lo-fi hip hop, 90 BPM, no vocals" \
  --user-id  {user_id} \
  --output-json

# DouBao BGM
python3 {baseDir}/scripts/ima_voice_create.py \
  --api-key  $IMA_API_KEY \
  --model-id  GenBGM \
  --prompt   "calm ambient piano background" \
  --user-id  {user_id} \
  --output-json

The script outputs JSON — parse it to get the result URL and pass it to the user via the UX protocol messages below.

Overview

Call IMA Open API to create AI-generated music/audio. All endpoints require an ima_* API key. The core flow is: query products → create task → poll until done.

🔒 Security & Transparency Policy

This skill is community-maintained and open for inspection.

🌐 Network Architecture

This skill uses a simpler network architecture than image/video skills:

Why the difference?

• Music generation (text_to_music) only needs text prompts → single API endpoint
• Image/video generation (i2i, i2v tasks) needs image file uploads → requires separate upload service

Security verification:

# Verify this skill only uses api.imastudio.com:
grep -n "https://" scripts/ima_voice_create.py

# Expected output:
# Only https://api.imastudio.com (no imapi.liveme.com)

✅ What Users CAN Do

Full transparency:

• ✅ Review all source code: Check scripts/ima_voice_create.py and ima_logger.py anytime
• ✅ Verify network calls: This skill uses only api.imastudio.com (music generation doesn't require image uploads). Verify by running: grep -n "https://" scripts/ima_voice_create.py
• ✅ Inspect local data: View ~/.openclaw/memory/ima_prefs.json and log files
• ✅ Control privacy: Delete preferences/logs anytime, or disable file writes (see below)

Configuration allowed:

• ✅ Set API key in environment or agent config:
  • Environment variable: export IMA_API_KEY=ima_your_key_here
  • OpenClaw/MCP config: Add IMA_API_KEY to agent's environment configuration
  • Get your key at: https://imastudio.com
• ✅ Use scoped/test keys: Test with limited API keys, rotate after testing
• ✅ Disable file writes: Make prefs/logs read-only or symlink to /dev/null

Data control:

• ✅ View stored data: cat ~/.openclaw/memory/ima_prefs.json
• ✅ Delete preferences: rm ~/.openclaw/memory/ima_prefs.json (resets to defaults)
• ✅ Delete logs: rm -rf ~/.openclaw/logs/ima_skills/ (auto-cleanup after 7 days anyway)

⚠️ Advanced Users: Fork & Modify

If you need to modify this skill for your use case:

1. Fork the repository (don't modify the original)
2. Update your fork with your changes
3. Test thoroughly with limited API keys
4. Document your changes for troubleshooting

Note: Modified skills may break API compatibility or introduce security issues. Official support only covers the unmodified version.

❌ What to AVOID (Security Risks)

Actions that could compromise security:

• ❌ Sharing API keys publicly or in skill files
• ❌ Modifying API endpoints to unknown servers
• ❌ Disabling SSL/TLS certificate verification
• ❌ Logging sensitive user data (prompts, IDs, etc.)
• ❌ Bypassing authentication or billing mechanisms

Why this matters:

1. API Compatibility: Skill logic aligns with IMA Open API schema
2. Security: Malicious modifications could leak credentials or bypass billing
3. Support: Modified skills may not be supported
4. Community: Breaking changes affect all users

📁 File System Access (Declared)

This skill reads/writes the following files:

What's stored:

• ✅ Model preferences (e.g., "last used: Suno sonic-v5")
• ✅ Timestamps (e.g., "2026-02-27 12:34:56")
• ✅ Task IDs and HTTP status codes
• ❌ NO API keys
• ❌ NO personal data
• ❌ NO prompts or generated content

Full transparency: See the complete data flow and privacy policy in the skill documentation above.

📋 Privacy & Data Handling Summary

What this skill does with your data:

Privacy recommendations:

1. Use test/scoped API keys for initial testing
2. Note: --user-id is never sent to IMA servers - it's only used locally as a key for storing preferences in ~/.openclaw/memory/ima_prefs.json
3. Review source code at scripts/ima_voice_create.py to verify network calls (search for create_task function)
4. Rotate API keys after testing or if compromised

Get your IMA API key: Visit https://imastudio.com to register and get started.

🔧 For Skill Maintainers Only

Version control:

• All changes must go through Git with proper version bumps (semver)
• CHANGELOG.md must document all changes
• Production deployments require code review

File checksums (optional):

# Verify skill integrity
sha256sum SKILL.md scripts/ima_voice_create.py

If users report issues, verify file integrity first.

🧠 User Preference Memory

User preferences override recommended defaults. If a user has generated before, use their preferred model — not the system default.

Storage: ~/.openclaw/memory/ima_prefs.json

{
  "user_{user_id}": {
    "text_to_music": { "model_id": "sonic", "model_name": "Suno", "credit": 25, "last_used": "..." }
  }
}

If the file or key doesn't exist, fall back to the ⭐ Recommended Defaults below.

When to Read (Before Every Generation)

1. Load ~/.openclaw/memory/ima_prefs.json (silently, no error if missing)
2. Look up user_{user_id}.text_to_music
3. If found → use that model; mention it:

   🎵 根据你的使用习惯，将用 [Model Name] 帮你生成音乐…
   • 模型：[Model Name]（你的常用模型）
   • 预计耗时：[X ~ Y 秒]
   • 消耗积分：[N pts]
   

4. If not found → use the ⭐ Recommended Default (Suno sonic-v5)

When to Write (After Every Successful Generation)

Save the used model to ~/.openclaw/memory/ima_prefs.json under user_{user_id}.text_to_music.
See ima-image-ai/SKILL.md → "User Preference Memory" for the full Python write snippet.

When to Update (User Explicitly Changes Model)

⭐ Recommended Defaults

These are fallback defaults — only used when no user preference exists.
Always default to the newest and most popular model. Do NOT default to the cheapest.

Selection guide by use case:

• Custom song with lyrics, vocals, style → Suno sonic-v5 (default)
• Background music / ambient loop → DouBao BGM
• Simple song generation → DouBao Song
• User explicitly asks for cheapest → DouBao BGM/Song (6pts each) — only if explicitly requested

⚠️ For Suno: model_version inside parameters (e.g. sonic-v5) is different from the outer model_version field (which is sonic). Always set both.

💬 User Experience Protocol (IM / Feishu / Discord) v1.1 🆕

v1.1 Update: Added Step 0 to ensure correct message ordering in group chats (learned from ima-image-ai v1.2).

Music generation completes in 10~45 seconds. Never let users wait in silence.
Always follow all 5 steps below, every single time.

🚫 Never Say to Users

Only tell users: model name · estimated time · credits · result (audio file/player) · plain-language status.

Estimated Generation Time per Model

estimated_max_seconds = upper bound (e.g. 45 for Suno, 25 for DouBao BGM/Song).

Step 0 — Initial Acknowledgment Reply (Normal Reply) 🆕

⚠️ CRITICAL: This step ensures correct message ordering in group chats.

Before doing anything else, reply to the user with a friendly acknowledgment message using your normal reply (not message tool). This reply will automatically appear FIRST in the conversation.

Example acknowledgment messages:

好的！马上帮你生成轻松的背景音乐 🎵

收到！用 Suno 为你创作一首歌 🎶

OK! Starting music generation with DouBao BGM 🎵

Rules:

• Keep it short and warm (< 15 words)
• Match the user's language (Chinese/English)
• Include relevant emoji (🎵/🎶/🎸/🎹)
• This is your ONLY normal reply — all subsequent updates use message tool

Why this matters:

• Normal replies automatically appear FIRST in the conversation thread
• message tool pushes appear in chronological order AFTER your initial reply
• This ensures users see: "好的！" → "🎵 开始生成..." → "✅ 成功!" (correct order)

Step 1 — Pre-Generation Notification (Push via message tool)

After Step 0 reply, use the message tool to push a notification immediately to the group/channel:

# Agent implementation
message(
    action="send",
    target=group_id,  # or channel_id for Discord
    message="""🎵 开始生成音乐，请稍候…
• 模型：[Model Name]
• 预计耗时：[X ~ Y 秒]
• 消耗积分：[N pts]"""
)

User-facing message template:

🎵 开始生成音乐，请稍候…
• 模型：[Model Name]
• 预计耗时：[X ~ Y 秒]
• 消耗积分：[N pts]

Cost transparency:

• Balanced (Suno 25 pts): "使用 Suno（25 积分，功能最全）"
• DouBao alternatives (30 pts each): "使用 DouBao BGM（30 积分）" — only if user explicitly requests DouBao or background music type

Adapt language to match the user. English → 🎵 Starting music generation, please wait [X~Y] seconds…

Step 2 — Progress Updates

Poll the task detail API every 5s.
Send a progress update every [Send Progress Every] seconds per the table above.

⏳ 音乐生成中… [P]%
已等待 [elapsed]s，预计最长 [max]s

Progress formula:

P = min(95, floor(elapsed_seconds / estimated_max_seconds * 100))

• Cap at 95% — never show 100% until the API returns success
• If elapsed > estimated_max: keep P at 95% and append 「快好了，稍等…」

Step 3 — Success Notification (Push audio via message tool)

When task status = success, use the message tool to send the generated audio directly (not as a text URL):

Agent implementation:

# Get result URL from script output or task detail API
result = get_task_result(task_id)
audio_url = result["medias"][0]["url"]

# Push audio + caption to group/channel
message(
    action="send",
    target=group_id,
    media=audio_url,  # Feishu/Discord will render the audio
    caption=f"""✅ 音乐生成成功！
• 模型：[Model Name]
• 耗时：预计 [X~Y]s，实际 [actual]s
• 消耗积分：[N pts]

🔗 原始链接：{audio_url}"""
)

User-facing message:

✅ 音乐生成成功！
• 模型：[Model Name]
• 耗时：预计 [X~Y]s，实际 [actual]s
• 消耗积分：[N pts]

🔗 原始链接：https://ws.esxscloud.com/.../audio.wav

[音频直接显示为文件卡片，可点击播放]

Platform-specific notes:

• Feishu: message(action=send, media=url, caption="...") — caption appears with audio file card
• Discord: Audio embeds automatically from URL; caption can be in message text
• Telegram: Use message(action=send, media=url, caption="...")

⚠️ Important:

• Always send audio via media parameter (file card/player) + include URL in caption text
• Do NOT use local file paths like /tmp/audio.wav — use HTTP URL from API
• Users expect: (1) clickable audio file card + (2) raw URL link for sharing/downloading
• Format: media=audio_url + caption="...🔗 原始链接：{audio_url}"

Step 4 — Failure Notification (Push via message tool)

When task status = failed or any API/network error, push a failure message with alternative suggestions:

Agent implementation:

message(
    action="send",
    target=group_id,
    message="""❌ 音乐生成失败
• 原因：[natural_language_error_message]
• 建议改用：
  - [Alt Model 1]（[特点]，[N pts]）
  - [Alt Model 2]（[特点]，[N pts]）

需要我帮你用其他模型重试吗？"""
)

⚠️ CRITICAL: Error Message Translation

NEVER show technical error messages to users. Always translate API errors into natural language.
API key & credits: 密钥与积分管理入口为 imaclaw.ai（与 imastudio.com 同属 IMA 平台）。Key and subscription management: imaclaw.ai (same IMA platform as imastudio.com).

Generic fallback (when error is unknown):

• Chinese: 音乐生成遇到问题，请稍后重试或换个模型试试
• English: Music generation encountered an issue, please try again or use another model

Best Practices:

1. Focus on user action: Tell users what to do next, not what went wrong technically
2. Be reassuring: Use phrases like "建议换个模型试试" instead of "生成失败了"
3. Avoid blame: Never say "你的描述有问题" → say "描述需要调整一下"
4. Provide alternatives: Always suggest 1-2 alternative models in the failure message
5. Music-specific:
   • For Suno lyrics errors, suggest simplifying lyrics or using auto-generated lyrics
   • For prompt length errors, give example length (e.g., "建议20-100字")
   • For BGM requests, recommend DouBao BGM over Suno
6. 🆕 Include actionable links (v1.0.8+): For 401/4008 errors, provide clickable links to API key generation or credit purchase pages

🆕 Enhanced Error Handling (v1.0.8):

Music generation uses direct error handling (no Reflection mechanism due to simpler parameters):

• 401 Unauthorized: System provides clickable link to API key generation page
• 4008 Insufficient Points: System provides clickable link to credit purchase page
• Other errors: Clear natural language explanations with alternative model suggestions

Error messages are user-friendly and actionable — users receive clear next steps for resolution.

Failure fallback table:

Step 5 — Done (No Further Action Needed) 🆕

v1.1 Note: After completing Steps 0-4:

• ✅ Step 0 already sent your normal reply (appears FIRST in chat)
• ✅ Steps 1-4 pushed all updates via message tool (appear in order)
• ✅ No further action needed — conversation is complete

Do NOT:

• ❌ Reply again with NO_REPLY (you already replied in Step 0)
• ❌ Send duplicate confirmation messages
• ❌ Use message tool to send the same content twice

Why this works:

User: "帮我生成一段轻松的背景音乐"
  ↓
[Step 0] Your normal reply:  "好的！马上帮你生成轻松的背景音乐 🎵"  ← Appears FIRST
  ↓
[Step 1] message tool push:  "🎵 开始生成音乐..."  ← Appears SECOND
  ↓
[Step 2] message tool push:  "⏳ 正在生成中… 45%"  ← (if task takes >15s)
  ↓
[Step 3] message tool push:  "✅ 音乐生成成功! [Audio File]"  ← Appears LAST
  ↓
[Step 5] Done. No further replies.

Supported Models

text_to_music (3 models)

Model guidance:

• Suno: Most powerful option. Supports full custom mode with genre tags, explicit instrumental toggle, vocal gender selection, and negative tags to exclude unwanted styles.
• DouBao BGM: Lightweight background music generation. Ideal for ambient / background tracks.
• DouBao Song: Song generation. Good for structured vocal compositions.

What you can generate:

• Background music (lo-fi, ambient, cinematic, electronic, jazz, classical…)
• Custom jingles or theme songs with specific BPM and key
• Vocal or instrumental tracks with mood direction
• Short loops or full-length compositions

Prompt writing tips (for Suno gpt_description_prompt):

• Genre: "lo-fi hip hop", "orchestral cinematic", "upbeat pop", "dark ambient"
• Tempo: "80 BPM", "fast tempo", "slow ballad"
• Vocals: "no vocals" → set make_instrumental=true; "female vocals" → vocal_gender="female"
• Mood: "happy and energetic", "melancholic", "tense and dramatic"
• Negative: negative_tags="heavy metal, distortion" to exclude styles
• Duration hint: "60 seconds", "30 second loop"

Environment

Base URL: https://api.imastudio.com

Required/recommended headers for all /open/v1/ endpoints:

Authorization: Bearer ima_your_api_key_here
x-app-source: ima_skills
x_app_language: en

⚠️ MANDATORY: Always Query Product List First

CRITICAL: You MUST call /open/v1/product/list BEFORE creating any task.
The attribute_id field is REQUIRED in the create request. If it is 0 or missing, you get:
"Invalid product attribute" → "Insufficient points" → task fails completely.
NEVER construct a create request from the model table alone. Always fetch the product first.

How to get attribute_id

# Step 1: Query product list
GET /open/v1/product/list?app=ima&platform=web&category=text_to_music

# Step 2: Walk the tree to find your model
for group in response["data"]:
    for version in group.get("children", []):
        if version["type"] == "3" and version["model_id"] == target_model_id:
            attribute_id  = version["credit_rules"][0]["attribute_id"]
            credit        = version["credit_rules"][0]["points"]
            model_version = version["id"]
            model_name    = version["name"]

Quick Reference: Known attribute_ids

⚠️ Production warning: attribute_id and credit values change frequently. Always call /open/v1/product/list at runtime; table below is pre-queried reference (2026-02-27).

Common Mistakes (and resulting errors)

Core Flow

1. GET /open/v1/product/list?app=ima&platform=web&category=text_to_music
   → REQUIRED: Get attribute_id, credit, model_version, form_config defaults

2. POST /open/v1/tasks/create
   → Must include: attribute_id, model_name, model_version, credit, cast, prompt (nested!)

3. POST /open/v1/tasks/detail  {task_id: "..."}
   → Poll every 3–5s until medias[].resource_status == 1
   → Extract url from completed media (mp3)

Supported Task Types

Detail API status values

Important: Treat resource_status: null as 0. Stop only when all medias have resource_status == 1. Check status != "failed" when rs=1.

API 1: Product List

GET /open/v1/product/list?app=ima&platform=web&category=text_to_music

Returns a V2 tree structure: type=2 nodes are model groups, type=3 nodes are versions (leaves). Only type=3 nodes contain credit_rules and form_config.

How to pick a version:

1. Traverse nodes to find type=3 leaves
2. Use model_id and id (= model_version) from the leaf
3. Pick credit_rules[].attribute_id
4. Use form_config[].value as default parameters values

API 2: Create Task

POST /open/v1/tasks/create

text_to_music

No image input. src_img_url: [], input_images: [].

{
  "task_type": "text_to_music",
  "enable_multi_model": false,
  "src_img_url": [],
  "parameters": [{
    "attribute_id":  "<from credit_rules>",
    "model_id":      "<model_id>",
    "model_name":    "<model_name>",
    "model_version": "<version_id>",
    "app":           "ima",
    "platform":      "web",
    "category":      "text_to_music",
    "credit":        "<points>",
    "parameters": {
      "prompt":       "upbeat electronic, 120 BPM, no vocals",
      "n":            1,
      "input_images": [],
      "cast":         {"points": "<points>", "attribute_id": "<attribute_id>"}
    }
  }]
}

Prompt tips for music generation:

• Genre: "upbeat electronic", "classical piano", "ambient chill"
• Tempo: "120 BPM", "slow tempo"
• Vocals: "no vocals", "male vocals", "female vocals"
• Mood: "happy", "melancholic", "energetic"
• Duration hint: "60 seconds", "short loop"

Key fields:

Response: data.id = task ID for polling.

API 3: Task Detail (Poll)

POST /open/v1/tasks/detail
{"task_id": "<id from create response>"}

Poll every 3–5s. Completed response:

{
  "id": "task_abc",
  "medias": [{
    "resource_status": 1,
    "url":          "https://cdn.../output.mp3",
    "duration_str": "60s",
    "format":       "mp3"
  }]
}

Output fields: url (mp3), duration_str, format.

Common Mistakes

Python Example

import time
import requests

BASE_URL = "https://api.imastudio.com"
API_KEY  = "ima_your_key_here"
HEADERS  = {
    "Authorization":  f"Bearer {API_KEY}",
    "Content-Type":   "application/json",
    "x-app-source":   "ima_skills",
    "x_app_language": "en",
}


def get_products(category: str) -> list:
    """Returns flat list of type=3 version nodes from V2 tree."""
    r = requests.get(
        f"{BASE_URL}/open/v1/product/list",
        headers=HEADERS,
        params={"app": "ima", "platform": "web", "category": category},
    )
    r.raise_for_status()
    nodes = r.json()["data"]
    versions = []
    for node in nodes:
        for child in node.get("children") or []:
            if child.get("type") == "3":
                versions.append(child)
            for gc in child.get("children") or []:
                if gc.get("type") == "3":
                    versions.append(gc)
    return versions


def create_music_task(prompt: str, product: dict) -> str:
    """Returns task_id."""
    rule = product["credit_rules"][0]
    form_defaults = {f["field"]: f["value"] for f in product.get("form_config", []) if f.get("value") is not None}

    nested_params = {
        "prompt": prompt,
        "n":      1,
        "input_images": [],
        "cast":   {"points": rule["points"], "attribute_id": rule["attribute_id"]},
        **form_defaults,
    }

    body = {
        "task_type":          "text_to_music",
        "enable_multi_model": False,
        "src_img_url":        [],
        "parameters": [{
            "attribute_id":  rule["attribute_id"],
            "model_id":      product["model_id"],
            "model_name":    product["name"],
            "model_version": product["id"],
            "app":           "ima",
            "platform":      "web",
            "category":      "text_to_music",
            "credit":        rule["points"],
            "parameters":    nested_params,
        }],
    }
    r = requests.post(f"{BASE_URL}/open/v1/tasks/create", headers=HEADERS, json=body)
    r.raise_for_status()
    return r.json()["data"]["id"]


def poll(task_id: str, interval: int = 3, timeout: int = 300) -> dict:
    deadline = time.time() + timeout
    while time.time() < deadline:
        r = requests.post(f"{BASE_URL}/open/v1/tasks/detail", headers=HEADERS, json={"task_id": task_id})
        r.raise_for_status()
        task   = r.json()["data"]
        medias = task.get("medias", [])
        if medias:
            if any(m.get("status") == "failed" for m in medias):
                raise RuntimeError(f"Task failed: {task_id}")
            rs = lambda m: m.get("resource_status") if m.get("resource_status") is not None else 0
            if any(rs(m) == 2 for m in medias):
                raise RuntimeError(f"Task failed: {task_id}")
            if all(rs(m) == 1 for m in medias):
                return task
        time.sleep(interval)
    raise TimeoutError(f"Task timed out: {task_id}")


# text_to_music
products = get_products("text_to_music")
task_id  = create_music_task("upbeat electronic, 120 BPM, no vocals", products[0])
result   = poll(task_id)
print(result["medias"][0]["url"])          # mp3 URL
print(result["medias"][0]["duration_str"]) # e.g. "60s"

Supported Models & Search Terms

Models: Suno sonic v4, Suno sonic v5, DouBao BGM (GenBGM), DouBao Song (GenSong)

Capabilities: music generation, text-to-music, AI music, background music, BGM, soundtrack, jingle, song with lyrics, vocal, instrumental, ambient music, audio generation