narrative-ai-framework 0.1.7

AI-powered voice diary framework: STT, TTS, LLM, RAG, and voice-agent engines

Narrative AI SDK

---

🔑 LLM Engine (nai.llm)

generate()

Description	Inputs	Returns
Standard text generation	prompt, model, max_tokens	LLMResponse

import narrative_ai as nai
import asyncio
async def main():
    nai.llm.set_api_key("key", provider="openai")
    res = await nai.llm.generate("Hi")
    print(res.text)
asyncio.run(main())

generate_stream()

Description	Inputs	Returns
Streaming responses	prompt, model	AsyncIterator

import narrative_ai as nai
import asyncio
async def main():
    nai.llm.set_api_key("key", provider="openai")
    async for chunk in nai.llm.generate_stream("Hi"):
        print(chunk)
asyncio.run(main())

set_api_key()

Description	Inputs	Returns
Set API key	api_key, provider	None

import narrative_ai as nai
nai.llm.set_api_key("key", "openai")

set_llm_provider()

Description	Inputs	Returns
Change provider	provider	None

import narrative_ai as nai
nai.llm.set_llm_provider("anthropic")

set_service_url()

Description	Inputs	Returns
Set base URL	url	None

import narrative_ai as nai
nai.llm.set_service_url("https://api.openai.com/v1")

---

🎙️ STT Engine (nai.stt)

transcribe()

Description	Inputs	Returns
Audio to text	audio_path, language	STTResult

import narrative_ai as nai
import asyncio
async def main():
    nai.stt.set_api_key("key", "elevenlabs")
    res = await nai.stt.transcribe("a.mp3")
    print(res.text)
asyncio.run(main())

stream_transcribe()

Description	Inputs	Returns
Real-time transcribing	audio_stream	AsyncIterator

import narrative_ai as nai
import asyncio
async def main():
    async for t in nai.stt.stream_transcribe(stream):
        print(t.text)
asyncio.run(main())

set_api_key()

Description	Inputs	Returns
Set STT key	key, provider	None

import narrative_ai as nai
nai.stt.set_api_key("key", "elevenlabs")

---

🔊 TTS Engine (nai.tts)

synthesize()

Description	Inputs	Returns
Text to audio	text, voice	str (Path)

import narrative_ai as nai
import asyncio
async def main():
    nai.tts.set_api_key("key", "openai")
    path = await nai.tts.synthesize("Hi")
    print(path)
asyncio.run(main())

stream_synthesize()

Description	Inputs	Returns
Stream audio bytes	text, voice	AsyncIterator

import narrative_ai as nai
import asyncio
async def main():
    async for chunk in nai.tts.stream_synthesize("Hi"):
        print(len(chunk))
asyncio.run(main())

---

📚 RAG Engine (nai.rag)

remember()

Description	Inputs	Returns
Index document	doc, doc_id	bool

import narrative_ai as nai
import asyncio
async def main():
    doc = await nai.input_processor.process("f.pdf")
    await nai.rag.remember(doc, "id_1")
asyncio.run(main())

recall()

Description	Inputs	Returns
Search context	query, top_k	RichContext

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.rag.recall("query")
    print(res.formatted_text)
asyncio.run(main())

forget()

Description	Inputs	Returns
Delete doc	doc_id	bool

import narrative_ai as nai
import asyncio
async def main():
    await nai.rag.forget("id_1")
asyncio.run(main())

clear_memory()

Description	Inputs	Returns
Wipe vector store	None	bool

import narrative_ai as nai
import asyncio
async def main():
    await nai.rag.clear_memory()
asyncio.run(main())

---

👁️ OCR Engine (nai.ocr)

process_image()

Description	Inputs	Returns
Image to text	image_path	OCRResult

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.ocr.process_image("i.jpg")
    print(res.text)
asyncio.run(main())

process_pdf()

Description	Inputs	Returns
PDF to text	pdf_path	OCRResult

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.ocr.process_pdf("d.pdf")
    print(res.text)
asyncio.run(main())

---

🛠️ Input Processor (nai.input_processor)

process()

Description	Inputs	Returns
Smart processing	source	StructuredDocument

import narrative_ai as nai
import asyncio
async def main():
    doc = await nai.input_processor.process("data.zip")
    print(doc.text)
asyncio.run(main())

process_batch()

Description	Inputs	Returns
Batch process	list	List[Doc]

import narrative_ai as nai
import asyncio
async def main():
    docs = await nai.input_processor.process_batch(["f1", "f2"])
asyncio.run(main())

process_audio()

Description	Inputs	Returns
Explicit audio	path	Doc

import narrative_ai as nai
import asyncio
async def main():
    doc = await nai.input_processor.process_audio("a.wav")
asyncio.run(main())

process_document()

Description	Inputs	Returns
Explicit PDF/DOCX	path	Doc

import narrative_ai as nai
import asyncio
async def main():
    doc = await nai.input_processor.process_document("d.pdf")
asyncio.run(main())

---

🤖 Voice Mode (nai.voice_mode)

start_agent()

Description	Inputs	Returns
Run agent loop	None	None

import narrative_ai as nai
nai.voice_mode.start_agent()

set_livekit_config()

Description	Inputs	Returns
Set connection	url, key, secret	None

import narrative_ai as nai
nai.voice_mode.set_livekit_config("url", "key", "secret")

set_agent_name()

Description	Inputs	Returns
Set name	name	None

import narrative_ai as nai
nai.voice_mode.set_agent_name("Assistant")

---

🔍 Web Intelligence (nai.web_intel)

search()

Description	Inputs	Returns
Real-time search	query	WebResult

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.web_intel.search("AI")
asyncio.run(main())

research()

Description	Inputs	Returns
Deep report	topic	str

import narrative_ai as nai
import asyncio
async def main():
    report = await nai.web_intel.research("Topic")
asyncio.run(main())

---

🎨 VLM Engine (nai.vlm)

analyze_image()

Description	Inputs	Returns
Image reasoning	image, prompt	VLMResponse

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.vlm.analyze_image("i.jpg", "What is this?")
asyncio.run(main())

chat_with_image()

Description	Inputs	Returns
Multi-turn chat	image, history	VLMResponse

import narrative_ai as nai
import asyncio
async def main():
    res = await nai.vlm.chat_with_image("i.jpg", history=[])
asyncio.run(main())

---

License

MIT License.