Written quotes within 60 minutes during business hours. Intake desk open 24/7. +1 (833) 282 8883 info@prismlinguistics.com

Transcription Services

Human transcription of recordings in any language, with translation and certification where the transcript is going to court, USCIS or a regulator.

Prism Language Solutions transcribes audio and video recordings in English and in more than 300 other languages, and translates the transcript into English where needed: recorded interviews, depositions, body-camera and jail calls, wiretaps, voicemails, focus groups, lectures and meetings. Transcripts for court are prepared in the forensic format used by legal transcriptionists, with the source language and the English side by side, line-numbered and certified. Work is priced per audio minute, quoted after we have heard a sample of the recording, and delivered as a Word or PDF document.

Last reviewed September 11, 2026 by Sarah J, Prism Language Solutions

Transcription is the service clients most underestimate, because a recording seems short. Twenty minutes of a wiretapped phone call in Dominican Spanish with two people talking over a television, then translated for a federal jury, is not a twenty-minute job; the professional rule of thumb is that a minute of forensic audio takes about an hour of work, before translation. A clean forty-minute interview in one language is quick. The difference is audio quality, number of speakers, language and what the transcript is for, and we ask to hear the recording before we quote.

  • Certified translations prepared to meet USCIS requirements
  • Human linguists only, matched to the subject
  • Fixed written quote within 60 minutes in business hours
  • One office in New York, linguists in every state

Audio and Video Transcription in English and Other Languages

A transcript is a written record of what was said, and the first decision is how literal it should be. Clean verbatim removes false starts, fillers and repetitions and is right for a lecture, a focus group, a podcast or a business meeting where the reader wants the content. Full verbatim keeps everything, including the um, the stutter, the interruption and the cough, and is what a court, a linguist or an investigator needs. Speakers are labeled, timestamps added at the interval you specify, and passages that cannot be made out are marked as unintelligible with the time code rather than guessed. Every transcript is produced by a native speaker of the language on the recording; we do not deliver machine output with a human skim, because the errors that matter, a negation missed, a name misheard, are the ones a skim does not catch.

Formats we handle include MP3, WAV, M4A, MP4, MOV, dictation files, body-camera exports and the audio track of a video deposition. Send the file through the quote form or a secure transfer link we provide; do not email a large or sensitive file.

Transcription and Translation of Foreign-Language Recordings

When the recording is in another language and the reader needs English, the work has two stages, always in that order: a native speaker transcribes the source language in full, then a translator renders that transcript into English, and the two are delivered together. Translating directly from audio, without a written source transcript, is faster and cheaper and we do not do it for anything evidentiary, because there is then no way to check what the translator heard. The two-column result lets a reader who knows the language verify any line against the original, and it is the format legal transcription bodies recommend.

The hard part is usually not the words but the variety. A recording in Guatemalan Spanish may contain Mam; a Nigerian call switches between English, Pidgin and Yoruba in one sentence; a Cantonese voicemail from Hong Kong is not Mandarin. We assign by the variety actually on the recording, which is one reason we ask to hear it first. Language pages such as Spanish, Arabic and Cantonese describe the dialect questions that decide who does the work.

Certified Transcripts for Court: Forensic Transcription Format

Recorded evidence in a language other than English, whether a wiretap, a jail call, a body-camera clip, a recorded interview or a voicemail, is presented in court through a transcript and translation, and the format is settled practice among legal transcriptionists. The document is a table: a line number, a speaker label (named where identity is established, otherwise Male Voice 1, Female Voice 2), the verbatim source-language transcription, and the English translation on the same line, with time codes and with unintelligible or inaudible passages marked as such rather than filled in. Nothing is cleaned up: the false starts, the slang and the profanity stay, because opposing counsel will have the recording too.

The certification is what makes it evidence. The transcript is delivered with a signed certification from the transcriber-translator stating their qualifications, that the transcription is a complete and accurate record of the recording, and that the translation is complete and accurate; it can be sworn as a declaration or affidavit where the court or the rules require. Federal Rule of Evidence 604 requires an interpreter to be qualified and to give an oath or affirmation to make a true translation, and states have their own rules for translated documents: Texas Rule of Evidence 1009, for example, admits a translation on a qualified translator’s affidavit served at least 45 days before trial, with objections due 15 days before. Our transcriber-translators are available to testify, and we keep a chain-of-custody record for the file from receipt to delivery. For live interpreting in the same matter see court interpreting; for translated exhibits generally, legal translation.

Transcription Turnaround and Cost per Audio Minute

Transcription is priced per minute of audio, not per hour of our time, so the quote is fixed once we have heard the file. As a guide, human English transcription in the US market runs from about $1.50 to $3 per audio minute for clear single-speaker audio, rising with multiple speakers, poor audio or full-verbatim formatting. Foreign-language transcription starts higher, and transcription plus translation is priced as the two stages combined, typically in the range of $8 to $15 per audio minute for a common language and clear audio, and more for forensic work on difficult recordings. Certification is included for court work; a notarized declaration, rush delivery and testimony are itemized separately. The ranges for our other services are on the pricing page.

Typical guide ranges; the written quote fixes the price after we hear the recording
WorkGuide price per audio minuteTypical turnaround
English transcription, clean or full verbatim, clear audioabout $1.50 to $31 to 3 business days per hour of audio
Foreign-language transcription, source language onlyabout $3 to $62 to 5 business days per hour of audio
Transcription plus English translation, two-columnabout $8 to $155 to 10 business days per hour of audio
Forensic transcription and translation, poor audio or multiple speakers, certifiedquoted after review; often $15 to $302 to 4 weeks per hour of audio

Turnaround scales with the length of the recording and, above all, with its quality. A two-hour wiretap in a noisy car cannot be rushed without errors, and we say so rather than promise a date we will miss. Send the recording, tell us the language, the number of speakers, what the transcript is for and when the hearing is, and the quote comes back with a delivery date within one business day. Other services for attorneys and investigators are listed under translation services and law firms.

Ready to start? Send the document or the booking details through the quote form, call +1 (833) 282 8883, or email info@prismlinguistics.com. You get a fixed written price within 60 minutes during business hours.

Questions we get asked

Do you transcribe from the audio or translate it directly?
For foreign-language recordings we always transcribe the source language first, then translate the transcript, and deliver both side by side. Direct translation from audio leaves nothing to check against and is not accepted as forensic practice.
Can the transcript be used in court?
Yes. Court transcripts are prepared in the forensic two-column format, line-numbered and time-coded, with a signed certification of accuracy from the transcriber-translator that can be sworn as a declaration or affidavit. The transcriber-translator can testify to the work if required. Admissibility is decided by the court under its own rules.
How long does transcription take?
A clear English recording is usually back within one to three business days per hour of audio. Foreign-language transcription with translation takes five to ten business days per hour of audio, and forensic work on poor recordings longer. We give a delivery date with the quote.
Why do you need to hear the recording before quoting?
Because price and turnaround depend on audio quality, number of speakers, dialect and format far more than on length. A sample tells us whether it is a one-hour job or a ten-hour one, and the quote we then give is fixed.

Sources for this page