What comes back from an English recording
Choose Hindi as the language and keep "Audio and text output" switched on, and the job returns four things you can download separately:
- Hindi audio
- An M4A file: the translated Hindi speech laid over the music, jingles and room sound of your original recording
- Hindi text
- The full Hindi translation in Devanagari, ready to paste into a document
- English transcript
- What was said in the original, so you can check any line of the Hindi against it
- Hindi subtitles
- SRT and VTT files with each Hindi line timed to the original speech
Switch "Audio and text output" off and the same file costs a fraction as much: you get the Hindi translation as text and subtitles, with no new voice. For a recording you only want to understand, that's usually enough.
Where English audio needs to become Hindi
An English induction recording for new shop-floor staff becomes a Hindi M4A that plays on the speaker in the training room, so the supervisor doesn't have to translate it live every week.
An NGO's 8-minute English recording on ORS and hand-washing is turned into Hindi audio for community radio and WhatsApp groups in Uttar Pradesh and Bihar.
A founder's English interview podcast gets a Hindi feed. The host and the guest each sound like themselves in Hindi, which keeps a two-person conversation easy to follow.
An English lecture on accounting standards, recorded on a phone, becomes Hindi text and subtitles for students from Hindi-medium schools who revise better in Hindi.
How the Hindi voice is made
There is no voice to pick. Up to 8 speakers in a file are told apart, and each one gets a Hindi voice copied from about 10 seconds of their own clean speech in the recording. Someone who speaks for less than about 3 seconds is given the copied voice of a similar-sounding speaker, or a stock voice matching theirs. Background music and noise are separated first, so they don't end up inside the copied voice, and they are mixed back under the Hindi speech at the end.
Hindi usually takes more syllables than English to say the same thing. Each Hindi line is placed where the English line was spoken and sped up by at most 1.15× when it runs long, so a fast English speaker can produce Hindi that sounds brisk. The article on how AI Hindi dubbing works explains the steps in Hindi.
Hindi that listeners actually follow
- Hindi verbs change with the speaker's gender (वह गया / वह गई). The translation follows each speaker's estimated gender from their voice, so a woman's lines come out in feminine forms. Check a few lines anyway; gender errors in Hindi translation lists what to look for.
- There is no setting for the register of the Hindi: you choose Hindi and get standard written Hindi. Technical terms can come out more formal than people say them in conversation, so read them with your listeners in mind. Formal or everyday Hindi shows how to judge it.
- If your audience thinks in lakh and crore, check figures like "2.5 million" in the Hindi text.
- Hindi output is in Devanagari. Roman-script Hindi (Hinglish written in English letters) isn't produced.
Files that work
- Audio: MP3, WAV, M4A, AAC, OGG, FLAC, WebM. Phone voice recorders usually save M4A, which works as it is.
- Length: up to 2 hours per file. Split longer recordings at a natural break.
- Video works too: for an English video, use the English to Hindi video translator, which returns a Hindi MP4.
- Android WhatsApp voice notes (.opus) and many call-recorder files (.amr) need converting to MP3 or M4A first.
Clear speech gives clear Hindi. A close microphone, one person speaking at a time and low background music matter more than the file format; recording settings for speech has specifics.
What it costs
Hindi audio costs 50 credits per minute (5¢, ₹4.25), billed per started minute with a 2-minute minimum. Hindi text and subtitles only cost 1 credit per minute, with a 5-credit minimum.
Hindi audio: 1,500 credits ($1.50, ₹127.50). Hindi text only: 30 credits (3¢, ₹2.55).
Hindi audio: 400 credits (40¢, ₹34).
Hindi text and subtitles: 90 credits (9¢, ₹7.65).
New accounts get 100 free credits when they sign in with Google: one Hindi audio file up to 2 minutes long, or 100 minutes of Hindi text. Paid top-ups open soon, charged in Indian rupees through Cashfree Payments (UPI, cards or netbanking), from ₹85 for 1,000 credits; until then, the 100 free credits are what you can use. You see the exact cost before you start, and failed jobs are refunded automatically. Full details are on the pricing page.
Using the tool on this page
- The spoken language is detected automatically; you select the output language — the spoken language itself for an untranslated transcript, or another language for a translation. Files can be up to 2 hours long.
- Audio is sent over HTTPS in small chunks and deleted with its transcripts within 30 minutes of delivery (how files are protected).
- You can download the translated audio (voice over your original music and effects) as M4A, a transcript and subtitles (SRT and VTT). Every output, step and limit is explained on the audio translator page.
Frequently asked questions
Can I get Hindi text from English audio without the Hindi voice?
Yes. Switch "Audio and text output" off, choose Hindi and press "Transcribe audio". You get the English transcript plus the Hindi translation and Hindi subtitles at 1 credit per minute.
Will the Hindi audio keep the background music of my recording?
Yes. Music and background sound are separated from the speech, kept, and mixed back under the Hindi voice in the M4A file.
Can I choose a male or female Hindi voice?
No. Each speaker's Hindi voice is copied from their own voice in the recording, and the Hindi grammar follows their estimated gender.
Can it translate English audio into Tamil, Telugu or other Indian languages?
No. Hindi is the only Indian language mydubly translates into. The 21 languages are listed on the audio translator page.