AI Dubbing

Automatically dub audio and video content into 100+ languages while preserving the original voice.

Overview

The AI Dubbing API translates and re-voices audio and video content, maintaining the speaker's original voice characteristics across languages. It supports automatic speaker detection, lip-sync adjustment for video, and quality scoring for every output.

100+ target languages

Voice preservation

Lip-sync adjustment

Batch processing

Speaker detection

Quality scoring

Quickstart

Dub an audio file into a new language in just a few lines of code.

from nur import NurClient
client = NurClient()
# Create a dubbing job
job = client.dubbing.create(
    file="podcast_episode.mp3",
    source_language="en",
    target_languages=["es", "fr", "de"],
    preserve_voice=True
)
print(f"Job {job.id} created, status: {job.status}")
print(f"Estimated completion: {job.estimated_completion}")

Create Dubbing Job

POST/v1/dubbing/create

Submit an audio or video file for dubbing into one or more target languages. You can provide either a file upload or a public URL to the source media.

Parameter	Type	Description
source_url	string	Public URL of the audio or video file to dub.
file	file	Audio or video file upload. Provide either file or source_url.
source_languageREQUIRED	string	ISO 639-1 language code of the source content.
target_languagesREQUIRED	string[]	Array of ISO 639-1 language codes for the dubbed outputs.
preserve_voice	boolean	Whether to clone the speaker's voice for dubbed output. Default: true.
lip_sync	boolean	Enable lip-sync adjustment for video files. Default: false.

# Dub from a URL with lip-sync enabled
job = client.dubbing.create(
    source_url="https://example.com/video.mp4",
    source_language="en",
    target_languages=["ja", "ko", "zh"],
    preserve_voice=True,
    lip_sync=True
)
print(f"Job ID: {job.id}")
print(f"Status: {job.status}")
print(f"Estimated: {job.estimated_completion}")

Response

{
  "id": "dub_abc123def456",
  "status": "pending",
  "source_language": "en",
  "target_languages": ["ja", "ko", "zh"],
  "created_at": "2025-01-15T10:30:00Z",
  "estimated_completion": "2025-01-15T10:35:00Z"
}

Get Dubbing Job

GET/v1/dubbing/{job_id}

Retrieve the status and results of a dubbing job. Once the status is completed, the response includes download URLs for each target language.

Parameter	Type	Description
job_idREQUIRED	string	The unique identifier of the dubbing job.

import time
# Poll until the job is complete
while True:
    job = client.dubbing.get("dub_abc123def456")
    print(f"Status: {job.status}")
    if job.status == "completed":
        for lang, output in job.outputs.items():
            print(f"  {lang}: {output.download_url}")
        print(f"Duration: {job.duration}s")
        print(f"Quality scores: {job.quality_scores}")
        break
    elif job.status == "failed":
        print("Dubbing failed")
        break
    time.sleep(5)

Response

{
  "id": "dub_abc123def456",
  "status": "completed",
  "outputs": {
    "ja": { "download_url": "https://cdn.nur.ai/dub/abc123/ja.mp4" },
    "ko": { "download_url": "https://cdn.nur.ai/dub/abc123/ko.mp4" },
    "zh": { "download_url": "https://cdn.nur.ai/dub/abc123/zh.mp4" }
  },
  "duration": 324.5,
  "quality_scores": {
    "ja": 0.94,
    "ko": 0.91,
    "zh": 0.93
  }
}

List Supported Languages

GET/v1/dubbing/languages

Retrieve the full list of supported source and target languages for dubbing. No parameters are required.

languages = client.dubbing.languages()
for lang in languages:
    print(f"{lang.code}: {lang.name} (source: {lang.as_source}, target: {lang.as_target})")

Response

{
  "languages": [
    { "code": "en", "name": "English", "as_source": true, "as_target": true },
    { "code": "es", "name": "Spanish", "as_source": true, "as_target": true },
    { "code": "ja", "name": "Japanese", "as_source": true, "as_target": true },
    { "code": "zh", "name": "Chinese", "as_source": true, "as_target": true }
  ]
}

Response Objects

Reference for the objects returned by the AI Dubbing endpoints.

DubbingJob (Create)

Field	Type	Description
id	string	Unique identifier for the dubbing job.
status	string	One of: pending, processing, completed, failed.
source_language	string	ISO 639-1 code of the source language.
target_languages	string[]	Array of target language codes.
created_at	string	ISO 8601 timestamp of job creation.
estimated_completion	string	ISO 8601 timestamp of estimated completion.

DubbingJob (Get - Completed)

Field	Type	Description
id	string	Unique identifier for the dubbing job.
status	string	One of: pending, processing, completed, failed.
outputs	object	Map of language code to { download_url } for each target.
duration	number	Duration of the source media in seconds.
quality_scores	object	Map of language code to quality score (0.0 - 1.0).

Best Practices

Use high-quality source audio

Clean source audio with minimal background noise produces significantly better dubbing results. Consider denoising with the Audio Tools API first.

Batch related languages together

Submit all target languages in a single request rather than creating separate jobs. This reduces processing time and ensures consistent voice cloning across outputs.

Check quality scores before publishing

The quality_scores field indicates how natural each dubbed output sounds. Scores below 0.85 may benefit from re-processing or manual review.

Use webhooks for long-running jobs

Instead of polling, configure a webhook URL in your dashboard to receive notifications when dubbing jobs complete. This is more efficient for production workflows.