Automatic Transcription

As a paid service, ATLAS.ti offers very fast, high-quality automatic transcription. Timestamps are accurate to the sentence. An hour of English audio is typically transcribed in under a minute.

When importing a single audio or video file, ATLAS.ti asks you whether you want to have the file transcribed automatically. You can also have a document transcribed automatically at a later time. Simply open it, then in the toolbar click Transcript and choose either Automatic Transcription….

Speaker Diarization and Coding

ATLAS.ti supports automatic speaker diarization. This will distinguish different speakers in the same audio track. Since ATLAS.ti does not know the names of the speakers, it will name them Speaker 1, Speaker 2, and so on—hence why we call this feature speaker diarization and not speaker detection.

Selecting "Code Speakers" here will create speaker codings for every spoken passage. This is very useful when you want to identify who said what via queries, smart codes, code cooccurrence analysis etc.

Before you go import the next file, rename the created speaker codes. ATLAS.ti will create a new Speaker 1 code when you transcribe the next audio if there isn't one, but it will re-use the Speaker 1 code if it is present. Since speakers are named in order of appearance per file, Speaker 1 in one document may be Speaker 2 in another, even if it is an interview with the same person.

Supported Languages

Automatic transcription supports over 30 languages, and bilingual audio for English and Spanish in the same track.

While ATLAS.ti supports automatic language detection, for best results we recommend selecting the language yourself.

  • Bulgarian
  • Catalan
  • Chinese
  • Chinese (Simplified)
  • Chinese (Simplified, People's Republic of China)
  • Chinese (Traditional)
  • Chinese (Traditional, Taiwan)
  • Czech
  • Danish
  • Dutch
  • English
  • English (Australia)
  • English (New Zealand)
  • English (India)
  • English (United Kingdom)
  • English (United States)
  • Estonian
  • Finnish
  • Flemish
  • French
  • French (Canada)
  • German
  • German (Switzerland)
  • Greek
  • Hindi
  • Hungarian
  • Indonesian
  • Italian
  • Japanese
  • Korean
  • Latvian
  • Lithuanian
  • Malay
  • Norwegian
  • Polish
  • Portuguese
  • Portuguese (Brazil)
  • Romanian
  • Russian
  • Slovak
  • Spanish
  • Spanish (Latin-America and Caribbean)
  • Swedish
  • Thai
  • Turkish
  • Ukrainian
  • Vietnamese

You can also import a transcript by clicking on Add Documents in the Home ribbon, then choosing Import Transcript.

Finally, if you have an audio or video document in ATLAS.ti already and you want to add a transcript, open the document, then in the toolbar click Transcript and choose either Automatic Transcription… or Import Transcript….

ATLAS.ti automatic transcriptions as well as some VTT files contain speaker information. Check the appropriate box, and ATLAS.ti will automatically insert speaker names into your text file, and code spoken passages with the appropriate speaker name, if the file contained speaker information.

See Working with Multimedia Transcripts for next steps.