Start transcription
In mocoVoice, you can start transcribing audio or video files directly from your web browser.
Steps
1. Log in to mocoVoice
Log in to mocoVoice and select Transcribe from the left sidebar.
2. Upload an audio or video file
Drag and drop files into the upload area in the center of the screen, or click the area to choose files. You can upload up to 6 audio or video files at once.
To upload multiple files, select and open multiple files in the file picker, or drag and drop multiple files into the upload area together.
When the upload finishes, the file names appear, and you can play each file to check it. Make sure the displayed file names match the audio or video files you want to transcribe.
3. Advanced settings (Optional)
Before you start transcribing, you can configure the following settings if needed.
Advanced settings are below the upload area. Scroll down slightly if you don’t see them.
When you select multiple files, a tab for each file appears above Advanced settings. Switch tabs to configure the title, meeting date, dictionary, languages, and other settings for each file.
- Title — The title for the transcript. If left blank, the file name is used. You can change this later on the result screen.
- Dictionary selection — Select a saved dictionary to improve recognition accuracy for technical terms and proper nouns.
- Languages in audio — If the audio mixes multiple languages, select all of them. The default is Japanese. If left blank, the AI will detect the language automatically. For supported languages, see Supported languages.
To select a language, search for it in the search bar and check the box. To let the AI decide, leave it on “Auto-detect” without selecting any languages.
4. Run transcription
Once your settings are ready, click Transcribe. When multiple files are selected, click Transcribe all to start transcribing all selected files together.
5. Processing
The system handles uploading and transcribing in order. You can check the progress in “Files” (transcript history). Processing will continue even if you close the page.
6. Check results
When finished, the screen switches to the results page. You will see the transcript text divided by speaker.
Speaker IDs in audio files over 3 hours
mocoVoice handles audio files longer than 3 hours, but the internal speaker separation process is split into 3-hour chunks. Because of this, speaker ID assignments may change at the 3-hour mark.
| Time | Speaker ID | Actual person |
|---|---|---|
| 0 to 3 hours | SPEAKER_00 | Person A |
| 0 to 3 hours | SPEAKER_01 | Person B |
| 3 to 6 hours | SPEAKER_02 | (Same) Person A |
| 3 to 6 hours | SPEAKER_03 | (Same) Person B |
If the same person appears after the 3-hour mark, they might get a new ID. Review your transcript and give the same person the same name to keep it organized (see Edit speaker names).
Actions on the result screen
Download results
Click the download button in the bottom right corner to save your transcript in these formats:
| Format | Extension | Description |
|---|---|---|
| Word | .docx | Microsoft Word format for general documents |
| Excel | .xlsx | Table format organizing timestamps and speakers |
| Markdown | .md | Lightweight plain text format with styling |
| Text | .txt | Simple plain text format |
| JSON | .json | Structured data for programming |
Edit names and results
If a single person receives multiple speaker IDs, assign them the same name to make the conversation easier to follow. For details, see Edit speaker names.
Create meeting notes
You can generate meeting notes from your transcripts of meetings or interviews. For details, see Create meeting notes.
Use dictionaries for better accuracy
If you want to improve recognition for technical terms or proper nouns, set up a dictionary beforehand. For details, see Improve accuracy with dictionaries.