Noiz.ai API Examples

Production-ready examples for integrating the Noiz.ai Text-to-Speech API into web apps, mobile clients, games, robots, and media pipelines. Use these samples to evaluate latency, voice quality, and multi-language workflows before you ship.

API overview

Endpoint	Method	Description
/v1/voices	GET	List voices (custom/built-in) with pagination and search
/v1/voices	POST	Clone a voice from an audio file or URL
/v1/voices/{voice_id}	GET	Get voice details
/v1/voices/{voice_id}	DELETE	Soft-delete a voice
/v1/text-to-speech	POST	Text-to-speech with emotion, speed, and format control
/v1/emotion-enhance	POST	Auto-annotate text with emotions for expressive speech

Base URL: https://noiz.ai/v1

Authentication: Send Authorization: <YOUR_API_KEY> on every request.

Repository layout

noizAPIExamples/
├── python/                            # Python examples
│   ├── basic/                         # Core API usage
│   ├── video_dubbing/                 # Video dubbing service
│   ├── game_dubbing/                  # Game NPC voice generation
│   ├── short_drama_translation/       # Short drama translation & dubbing
│   ├── robot_voice/                   # Robot voice integration
│   ├── smart_assistant/               # Smart assistant / IoT voice
│   ├── audiobook/                     # Audiobook production
│   ├── education/                     # Education & language learning
│   ├── ivr_system/                    # IVR phone system
│   └── content_creation/              # Short video & content creation
│
├── android/                           # Android (Kotlin) demo app
│   └── NoizTTSDemo/
│
├── javascript/                        # Node.js examples
│   ├── basic/
│   └── streaming/                     # Real-time TTS HTTP server
│
├── curl/                              # cURL command examples
│   └── examples.sh
│
└── csharp/                            # C# / Unity examples
    └── UnityGameDubbing/

Quick start

1. Get an API key

Sign up at the Noiz.ai developer portal and create an API key.

2. Set your environment

export NOIZ_API_KEY="your_api_key_here"

3. Python

cd python
pip install -r requirements.txt

python basic/text_to_speech.py
python basic/voice_clone.py
python video_dubbing/video_dubbing_service.py
python game_dubbing/game_npc_voice.py

4. Node.js

cd javascript
npm install

node basic/text_to_speech.js
node streaming/realtime_tts_server.js   # TTS server on port 3100

5. Android

Open android/NoizTTSDemo in Android Studio and add to local.properties:

noiz.api.key=your_api_key_here

6. cURL

export NOIZ_API_KEY="your_key"
bash curl/examples.sh

Industry use cases

Use case	Description	Path
Video dubbing	Dubbing with subtitle timing and duration control	python/video_dubbing/
Game NPC voices	Character voices with emotion for dialogue and barks	python/game_dubbing/
Short drama / localization	Cross-language dubbing while preserving voice character	python/short_drama_translation/
Robotics	Service, education, and companion robots with caching and queues	python/robot_voice/
Smart assistant	IoT and smart-home voice output	python/smart_assistant/
Audiobooks	Long-form narration with chapter splitting	python/audiobook/
EdTech	Vocabulary drills, dialogue practice, course narration	python/education/
IVR	Phone menus and dynamic status messages	python/ivr_system/
Content creation	Short-form video narration and batch voiceover	python/content_creation/
Unity games	In-game TTS with caching and NPC dialogue	csharp/UnityGameDubbing/

Core API capabilities

Text-to-speech

• Up to 5,000 characters per request
• Output: WAV / MP3
• Speed control (speed)
• Emotion control (emo, e.g. {"Sadness":0.2, "Surprise":0.5})
• Multi-language (zh, en, zh+en, and others per API)
• Duration control, silence trimming, similarity enhancement

Voice cloning

• Upload WAV / MP3 / M4A or pass a remote URL
• Optional denoising
• Multi-language cloning

Emotion enhancement

• Returns richer text for more expressive TTS

Security and operations

• Do not ship API keys in mobile or browser clients; keep keys on your server or use a backend proxy.
• Prefer environment variables or a secrets manager in CI/CD.
• TTS responses are binary audio streams; handle Content-Type and headers such as X-Audio-Duration as documented.
• Respect rate limits; add backoff between batch jobs.

Links

• Noiz.ai
• API documentation
• Developer portal