NVIDIA Nemotron 3.5 ASR API | Multilingual Speech To Text

NVIDIA Nemotron 3.5 ASR

NVIDIA Nemotron 3.5 ASR is a multilingual speech-to-text model that transcribes uploaded audio into text with automatic language detection. It is designed for audio transcription, podcast processing, subtitle preparation, meeting notes, voice analytics, localization workflows, and other production-ready ASR use cases.

Why Choose This?

Multilingual speech recognition
Transcribe spoken audio into text across multiple languages.
Automatic language detection
Leave language on auto when you want the model to detect the spoken language automatically.
Simple transcription workflow
Upload one audio file, optionally choose a language, and generate the transcript.
Useful for content and productivity workflows
Suitable for podcasts, subtitles, interviews, lectures, meetings, and other spoken-content pipelines.
Production-ready API
Easy to integrate into transcription, indexing, search, and accessibility workflows.

Parameters

Parameter	Required	Description
audio	Yes	Input audio file to transcribe.
language	No	Language setting for transcription. Use `auto` for automatic detection.

How to Use

Upload your audio — provide the audio clip you want to transcribe.
Choose language (optional) — leave it as auto for automatic detection, or select a fixed language if needed.
Submit — run the model and get the transcription result.

Example Output

{
  "output": "Ah, the garden is blooming again just like life with a little patience and care, everything finds its time to shine."
}

Pricing

Pricing is billed per started minute of input audio.

Audio Duration	Cost
1s–60s	$0.008
61s–120s	$0.016
121s–180s	$0.024

Billing Rules

Pricing is $0.008 per started minute
Audio duration is billed in started 60-second units
Audio shorter than 60 seconds is billed as 1 minute
language does not affect pricing

Best Use Cases

Audio transcription — Convert speech recordings into text.
Podcast processing — Generate transcripts for episodes and clips.
Subtitle preparation — Extract spoken content before caption formatting.
Meeting notes — Turn recorded discussions into readable text.
Voice analytics — Prepare transcripts for search, tagging, or downstream analysis.

Pro Tips

Use clean audio for better transcription accuracy.
Leave language as auto when the spoken language is clear and consistent.
Set a specific language when automatic detection may be ambiguous.
Short clips are useful for fast testing before processing longer recordings.
Review the output before publishing if the audio contains names, jargon, or strong accents.

Notes

audio is required.
language is optional.
Pricing is based on input audio duration and billed per started minute.
Better audio quality generally improves transcription quality.

Related Models

Other NVIDIA speech and multimodal workflows — Useful when you need text, audio, video, or vision processing beyond ASR.
Text-to-speech workflows — Useful when you need voice generation instead of transcription.
Subtitle and caption workflows — Useful when you need styled subtitle output rather than plain transcript text.

Nemotron 3.5 Asr API — Quick start

Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/nvidia/nemotron-3.5-asr with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read URLs from data.outputs. Examples for Nemotron 3.5 Asr below.

HTTP example

# Submit the prediction
curl --fail-with-body --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/nvidia/nemotron-3.5-asr" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY" \
  -d '{
    "audio": "https://example.com/your-audio.mp3",
    "language": "auto"
}'

# Wait at least 2 seconds, then poll. Safe GET requests may be retried.
curl --fail-with-body --connect-timeout 10 --max-time 30 \
  --retry 4 --retry-all-errors --retry-delay 1 \
  -X GET "https://api.wavespeed.ai/api/v3/predictions/{request_id}/result" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY"

# Start at 2 seconds and increase the interval for long-running tasks.
# Stop on completed, failed, cancelled, or timeout.

Node.js example

// npm install wavespeed
const { Client } = require('wavespeed');

const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');
const client = new Client(apiKey);

try {
  const result = await client.run("nvidia/nemotron-3.5-asr", {
        "audio": "https://example.com/your-audio.mp3",
        "language": "auto"
}, {
    timeout: 3600,
    pollInterval: 2.0,
  });
  console.log(result.outputs);
} catch (error) {
  console.error('Generation failed:', error);
  process.exitCode = 1;
}

Python example

# pip install wavespeed
import os
from wavespeed import Client

client = Client(api_key=os.environ["WAVESPEED_API_KEY"])

try:
    output = client.run(
        "nvidia/nemotron-3.5-asr",
        {
    "audio": "https://example.com/your-audio.mp3",
    "language": "auto"
},
        timeout=3600.0,
        poll_interval=2.0,
    )
    print(output["outputs"])
except Exception as error:
    raise SystemExit(f"Generation failed: {error}") from error

Nemotron 3.5 Asr API — Frequently asked questions

What is the Nemotron 3.5 Asr API?

Nemotron 3.5 Asr is a NVIDIA model for AI inference, exposed as a REST API on WaveSpeedAI. NVIDIA Nemotron 3.5 ASR is a fast AI speech-to-text model that transcribes multilingual audio into text with automatic language detection. Ready-to-use REST inference API for audio transcription, podcast processing, video subtitles, meeting notes, voice analytics, content localization, and professional ASR workflows with simple integration, no coldstarts, and affordable pricing. You can call it programmatically or try it from the playground above.

How do I call the Nemotron 3.5 Asr API?

POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. The playground generates production-oriented Python, JavaScript, and cURL examples with timeouts, transient-error handling, and safe GET retries. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/nvidia/nvidia-nemotron-3.5-asr.

How much does Nemotron 3.5 Asr cost per run?

Nemotron 3.5 Asr starts at $0.008 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.

What inputs does Nemotron 3.5 Asr accept?

Key inputs: `audio`, `language`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/nvidia/nvidia-nemotron-3.5-asr.

How do I get started with the Nemotron 3.5 Asr API?

Sign up for a free WaveSpeedAI account to claim starter credits, copy your API key from /accesskey, then call the endpoint shown in the API tab of the playground. The playground also auto-generates a code sample in Python, JavaScript, or cURL for the parameters you've set.

Can I use Nemotron 3.5 Asr outputs commercially?

Commercial usage rights depend on the model's license, set by its provider (NVIDIA). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.

ContohLihat semua

Model Terkait

README