Dubverse pricing, plans and limits

Last updated: 2026-10-11

Dubverse, founded in India in 2021, is a web platform that dubs video, adds subtitles and generates text-to-speech voiceovers, with the vendor claiming 60+ languages. It bills in credits, sells a separate text-to-speech API, and lets editors fix translations line by line, aimed at informational content rather than films.

About Dubverse

Dubverse is a browser-based AI dubbing, subtitling and text-to-speech platform made by Lensmatic Solutions Private Limited, an India-registered company (registered office in Greater Noida, per its privacy policy) that Varshul Gupta and Anuja Dhawan started in 2021. Inc42 lists one seed round of about $800K dated June 2022 with Kalaari Capital as the investor, and the vendor says the JioGenNext accelerator selected it early on. On 8 April 2026 Exotel acqui-hired the core team, including both co-founders, and Dubverse says the product keeps running on its own with no change to subscriptions or pricing. Its text-to-speech side overlaps AI voice assistants, though the main job is working on existing footage rather than generating presenters, much as Rask AI does.

You upload a video or paste a link, choose target languages and a stock speaker, and the platform transcribes, translates and voices the result. The editor puts text, audio and video on one screen, so you can rewrite a line, pick between pronunciations of a word, run find and replace across the script and assign different lines to different speakers in a mode called Retune. Translation is labelled Hawk (GPT-3.5) on Pro and Eagle (GPT-4) on Supreme, and the vendor says background sound effects survive the dub. Scripts that mix languages, such as Hinglish, are handled in one pass.

The credit system sets the real price. A minute of dubbing costs 4 credits, so, if every credit goes to dubbing, the smallest bundle covers about 12 minutes of video: roughly $1.44 per minute on monthly Pro billing and $0.72 on half-yearly billing. A bundle twenty times larger brings Pro to about $0.60 per minute monthly and $0.30 half-yearly. Subtitling and text-to-speech use fewer credits per minute, and a separate API bills text-to-speech by the character.

It suits course creators, training teams and marketers localizing explainers, software walkthroughs and support videos, which is the content the vendor itself recommends. The vendor advises against using it for films or other entertainment content, and recommends human review of every output. For voice-only work with no video, try ElevenLabs for narration or Murf AI for scripted voiceovers. HeyGen suits teams that want avatar presenters beside translation. The dubbing roundup, the text-to-speech guide and the translation guide compare the field. Teams that only need cloned voices have narrower options in Fish Audio or Resemble AI, while VEED covers browser editing around a dub and Synthesia builds avatar-led training videos. Lip sync is worth testing separately, and the lip-sync app guide shows how the options differ; the video generator guide and the video editing guide cover the neighbouring tools.

Outside opinion is mixed. G2 scores it well above Trustpilot, so read the recent complaints about voice and lip-sync quality before paying. The vendor's own pages also disagree with each other in places: the FAQ still says no API exists, while the docs describe a text-to-speech API with its own price list.

Pricing

Credits are spent per task: dubbing, subtitling and text-to-speech each draw a different amount, and a 2-day trial needs no card. On monthly billing, Pro is $18 and Supreme is $30 for 50 credits, and the plan slider scales to 1,000 credits at $150 and $250. Half-yearly billing shows $9 and $15 a month for 50 credits (billed $54 and $90 per six months), and 1,000 Pro credits at $75 a month.

Enterprise is quoted. The text-to-speech API is billed per character instead. Read in USD on 2026-10-11; the vendor also sells in rupees.

Per-minute cost is worth lining up beside Rask AI, whose plan sizes differ.

Key Features

• Line-by-line editing: Text, audio and video sit on one screen with find and replace, transliteration, a glossary and several pronunciation choices per word, which the vendor says cuts edit time by 50%.
• Retune for multiple speakers: Assign different lines of a dubbed video to different AI speakers, while automatic speaker detection is still labelled beta in the vendor FAQ.
• Voice cloning on Supreme: The Supreme card lists voice cloning so a branded voice can speak dubbed languages, though the FAQ describes clones as set up on request.
• Three text-to-speech models: neo.one favors speed, candy.two favors fidelity and emotion control, and swift.call adds runtime voice cloning, word timestamps and streaming for Hindi and English voicebots.
• Background audio kept: Sound effects in the source video stay under the new voice track, according to the vendor FAQ.
• Mixed-language scripts: Hinglish, Spanglish and other mixed scripts can be voiced in a single pass without splitting the text.
• Subtitles and exports: Automatic subtitles come with basic animation on Pro and custom styling on Supreme, plus VTT and XML download options.
• Low first-byte latency: The vendor's docs report an average 193 ms to first byte on neo.one, plus audio streaming and adjustable bitrate.

Pros

• Cost per dubbed minute falls sharply as the credit bundle grows, which suits teams with a steady backlog of explainers and training videos.
• G2 reviewers praise how fast existing video becomes multilingual, and an August 2026 review calls the voices natural enough for marketing and informational content.
• A separate pay-per-character API lets developers test voices before paying anything for a video plan.
• A GDPR-based data protection addendum is published inside the privacy policy, which gives European buyers something concrete to read.

Cons

• Quality is uneven outside the core languages: a G2 reviewer in August 2026 reported voice quality and pacing vary by language and wanted finer word-level controls, and a 2024 Trustpilot reviewer called the lip sync poor.
• Vendor pages contradict each other on voice count (200+ versus 500+), language count (the FAQ claims more than the 32 on its own language list) and API availability, so confirm your language before paying.
• The FAQ says voice cloning is not included in any plan and is billed separately, while the Supreme card lists it; its pricing details list lip sync only under Enterprise.
• In the vendor's own latency table, which cites a 2024 Artificial Analysis report, neo.one's 276 characters per second trails Amazon Polly, LMNT, Azure and Google Studio voices, so speed is not its selling point.
• The co-founders and core team moved to Exotel, so the product now depends on the remaining team; prefer the monthly plan until its roadmap is clearer.

Visit Dubverse Official Website