SDKs

Official client libraries for Python, JavaScript/TypeScript, and Go.

Python

Requires Python 3.8+. Supports both sync and async usage.

1pip install nur-ai

from nur import NurClient
client = NurClient()
# Text to Speech
audio = client.tts.generate(
    text="Hello from Python!",
    voice_id="rachel_v2"
)
audio.save("output.mp3")
# Speech to Text
transcript = client.stt.transcribe(file="audio.mp3")
print(transcript.text)
# Async usage
import asyncio
async def main():
    async with client.stt.stream(language="en") as stream:
        async for result in stream:
            print(result.text)
asyncio.run(main())

JavaScript / TypeScript

Works with Node.js 18+, Deno, and Bun. Full TypeScript types included.

1npm install @nur/sdk

import { NurClient } from '@nur/sdk';
const client = new NurClient();
// Text to Speech
const audio = await client.tts.generate({
  text: 'Hello from JavaScript!',
  voiceId: 'rachel_v2',
});
await audio.save('output.mp3');
// Speech to Text
const transcript = await client.stt.transcribe({
  file: 'audio.mp3',
});
console.log(transcript.text);
// Streaming
const stream = await client.tts.stream({
  text: 'Streaming audio...',
  voiceId: 'rachel_v2',
});
for await (const chunk of stream) {
  process.stdout.write(chunk);
}

Go

Requires Go 1.21+. Fully idiomatic with context support.

1go get github.com/nur-ai/nur-go

package main
import (
    "context"
    "fmt"
    nur "github.com/nur-ai/nur-go"
)
func main() {
    client := nur.NewClient()
    // Text to Speech
    audio, _ := client.TTS.Generate(context.Background(), &nur.TTSRequest{
        Text:    "Hello from Go!",
        VoiceID: "rachel_v2",
    })
    audio.Save("output.mp3")
    // Speech to Text
    transcript, _ := client.STT.Transcribe(context.Background(), &nur.STTRequest{
        File: "audio.mp3",
    })
    fmt.Println(transcript.Text)
}

Configuration

All SDKs support the following configuration options:

Option	Env Variable	Default	Description
api_key	NUR_API_KEY	-	Your API key
base_url	NUR_BASE_URL	https://api.nur.ai	API base URL
timeout	NUR_TIMEOUT	30s	Request timeout
max_retries	NUR_MAX_RETRIES	3	Max retry attempts for failed requests

Error Handling

All SDKs throw typed errors that you can catch and handle.

from nur import NurClient, NurError, RateLimitError
client = NurClient()
try:
    audio = client.tts.generate(text="Hello", voice_id="rachel_v2")
except RateLimitError as e:
    print(f"Rate limited. Retry after {e.retry_after}s")
except NurError as e:
    print(f"API error {e.status_code}: {e.message}")