NVIDIA-Nemotron-3.5-Transcribe-30B-A4B

# Nemotron 3.5 Transcribe runs on your side (here behind vLLM with the NeMo Speech
# plugin, one H100 80 GB per the model card). Its transcript is upserted into a
# collection whose text vector index makes the recording searchable.
import base64
from openai import OpenAI
import requests

asr = OpenAI(base_url="http://127.0.0.1:8000/v1", api_key="EMPTY")
audio = base64.b64encode(open("standup-2026-11-06.wav", "rb").read()).decode()
transcript = asr.chat.completions.create(
    model="nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B",
    messages=[{"role": "user", "content": [
        {"type": "text", "text": "Audio transcription. Language: as spoken. Wording: omit fillers and accidental repetitions; keep the remaining spoken wording unchanged. Formatting: punctuation and capitalization. Output: transcript only."},
        {"type": "audio_url", "audio_url": {"url": f"data:audio/wav;base64,{audio}"}},
    ]}],
    temperature=0,
    extra_body={"chat_template_kwargs": {"enable_thinking": False}, "skip_special_tokens": False},
).choices[0].message.content

requests.post(
    "https://api.mixpeek.com/v1/namespaces/ns_your_namespace/documents/upsert",
    headers={"Authorization": "Bearer API_KEY"},
    json={"collection_id": "col_your_collection",
          "documents": [{"document_id": "standup-2026-11-06",
                         "vectors": {"text-embedding": your_embedding_of(transcript)},
                         "payload": {"transcript": transcript, "source": "s3://recordings/standup-2026-11-06.wav"}}]},
)