Metadata-Version: 2.5
Name: gandr-langchain
Version: 0.1.0
Summary: LangChain text to speech tool for the Gandr TTS API. First audio byte in 146 ms over the open internet, 116 ms p50 first audio, server side warm. 23 languages, every render watermarked.
Project-URL: Homepage, https://gandr.ai
Project-URL: Documentation, https://gandr.ai/docs
Project-URL: Source, https://github.com/Gandr-AI/gandr-langchain
Author-email: Sam Alghaithi <sam@gandr.ai>
License: MIT
Keywords: gandr,langchain,text-to-speech,tts,voice-agents
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.9
Requires-Dist: gandr>=0.1.2
Requires-Dist: langchain-core>=0.3
Requires-Dist: pydantic>=2
Description-Content-Type: text/markdown

# gandr-langchain

LangChain text to speech tool for the [Gandr TTS API](https://gandr.ai), built for voice agents.

- First audio byte in **146 ms over the open internet, 116 ms p50 first audio, server side warm**
- Reads numbers, dates, order IDs and addresses correctly
- One engine speaks **23 languages** with six voices
- **Every render watermarked** (imperceptible, detectable)
- **$10 a month for one million tokens**, or unlimited, unmetered stream plans from **$150/mo** (annual)

Free key starts at **100,000 tokens, no card**: [gandr.ai](https://gandr.ai)

## Install

```bash
pip install gandr-langchain
```

## Use

```python
from gandr_langchain import GandrText2SpeechTool

tool = GandrText2SpeechTool()          # reads GANDR_API_KEY from the environment
path = tool.run("Order number 4-2-7-1 ships on March 3rd.")
# path is a WAV file of the spoken line

# pick a voice or a language per tool instance
tool = GandrText2SpeechTool(voice="gandr-leo", language="es")
```

Drops into any LangChain agent as a standard tool: the model passes the text to speak,
gets back the audio file path.

Docs: [gandr.ai/docs](https://gandr.ai/docs) · Voices: `gandr-ava`, `gandr-dane`, `gandr-jenny`, `gandr-leo`, `gandr-lewis`, `gandr-mia`
