To transcribe Spanish audio to text, upload it to Word's Transcribe feature, use an online transcription service, run the open-source Whisper on your computer or use a desktop app that transcribes locally. For client recordings, the deciding question is whether the audio leaves your computer. To get English, translate the Spanish transcript afterwards.
Four ways to transcribe Spanish audio
The four ways to transcribe Spanish audio differ in where the recording is processed, what they cost and how much setup they need. As of 29 October 2026:
| Where the audio goes | Cost | Effort | |
|---|---|---|---|
| Word Transcribe | Microsoft's cloud | Included with Microsoft 365: 300 minutes of uploads a month | Upload the file in Word |
| Online service (for example, Otter) | The provider's cloud | Free plan with 300 minutes a month and 30 minutes per conversation; paid plans for more | Upload the file |
| Whisper, open source | Your computer | Free (MIT licence) | Install Python and ffmpeg, use the command line |
| Desktop app that transcribes locally (for example, ablalo) | Your computer | Subscription after a free trial | Drag the file into the app |
Word: Transcribe
Word's Transcribe turns an uploaded recording into a transcript, and Spanish is supported, including Spanish (Spain), Spanish (Mexico) and Spanish (USA). Microsoft gives Microsoft 365 subscribers 300 minutes of uploaded audio a month, and 30,000 minutes with a Microsoft Copilot licence.
- Open a document and go to Home.
- Open the menu of the Dictate button and choose Transcribe.
- Click Upload audio and choose the file (.wav, .mp4, .m4a or .mp3).
There is a catch: Microsoft says Transcribe is currently only available in Word for Microsoft 365 on Windows for business (commercial) accounts, and that government accounts can use it only in Word for the web. The recording is processed in Microsoft's cloud.
Online transcription services
Online transcription services take an uploaded file or join your meetings and return a transcript and a summary. Otter, for example, transcribes Spanish among six languages; as of 29 October 2026 its free plan includes 300 minutes a month with a maximum of 30 minutes per conversation, and its Pro plan 1,200 minutes a month.
They are fast and need no setup. The trade-off is that the recording and the transcript live in the provider's cloud, under its retention terms. Before you upload a client's recording, check where it is stored, for how long and whether it is used to train the provider's models.
Whisper on your own computer
Whisper is an open-source speech recognition model, released under the MIT licence, that runs on your own computer. It is multilingual, so it handles Spanish, and it can also translate speech into English with its translate task, using one of its multilingual models.
It is free and nothing is uploaded, but it is a developer tool: you install it with Python and ffmpeg and run it from the command line. It suits technical users who transcribe now and then and are comfortable with that setup.
A desktop app that transcribes locally
A desktop app that transcribes locally combines Whisper's privacy with the convenience of an online service: you drag the file in, and the audio never leaves your computer.
With ablalo, you drop a Spanish recording into its Hearings tab (MP4, MOV, WebM, MKV, M4A, MP3, WAV, OGG, Opus, FLAC or AAC) and it is transcribed on your own computer, without uploading the file. It separates the speakers so you can name them, syncs the text with the video, lets you mark key passages with the exact second and exports to Word with timestamps. It can also record a Teams, Zoom or Meet call with your computer's sound, without a bot joining.
How to get the transcript in English
To turn Spanish audio into English text, the most reliable route is two steps: transcribe in Spanish, check the transcript against the recording, and then translate the corrected text. Translating straight from audio skips the check, and any word misheard in Spanish ends up mistranslated in English.
- With Whisper, the translate task produces English directly from the audio.
- With any other option, you translate the Spanish transcript with the translation tool you trust.
- With ablalo, the transcript of a recording comes out in Spanish, and you can translate passages of it with the AI Assistant. For your own dictation, «Always translate to English» writes English as you speak Spanish.
A machine translation is a working draft. If the English version is going to be filed or relied on by someone else, treat it like any other translation and have it reviewed.
Tips for a better Spanish transcript
The quality of a Spanish transcript depends more on the recording than on the tool. Four habits make the biggest difference:
- Record close to the speakers. A microphone in the middle of the table beats a laptop at the end of it.
- One person at a time. When two people talk over each other, every tool loses both sentences.
- Teach it the names. If the tool has a custom dictionary, add the names of the parties and companies before you start.
- Check numbers, dates and names against the audio before you quote them. Timestamps make that quick.
In short
To transcribe Spanish audio to text, you can use Word's Transcribe (Microsoft 365, 300 minutes a month, business accounts on Windows), an online service such as Otter, the open-source Whisper on your computer or a desktop app that transcribes locally, such as ablalo. For client recordings, prefer an option that keeps the audio on your computer, and to get English, translate the checked Spanish transcript.
Sources: Microsoft, Transcribe your recordings; Otter, pricing and plans; OpenAI, Whisper on GitHub. Checked on 29 October 2026.
Frequently asked questions
Can Word transcribe Spanish audio?
Yes. Word's Transcribe supports Spanish (Spain), Spanish (Mexico) and Spanish (USA), with 300 minutes of uploads a month for Microsoft 365 subscribers. Microsoft says it is available in Word for Microsoft 365 on Windows for business accounts.
Is there a free way to transcribe Spanish audio?
Yes. Whisper is free and open source and runs on your computer, but it needs a technical setup. Some online services have free plans with monthly limits, such as Otter's 300 minutes a month.
How do I translate Spanish audio to English text?
Transcribe the audio in Spanish, check the transcript against the recording and translate the corrected text. Whisper can also translate straight from Spanish audio to English text.
Can I transcribe a Spanish recording without uploading it?
Yes, with a tool that transcribes on your own computer: Whisper, or a desktop app such as ablalo, which transcribes the file locally and exports it to Word with timestamps.
Which audio formats can I transcribe?
Word Transcribe accepts .wav, .mp4, .m4a and .mp3. ablalo accepts MP4, MOV, WebM, MKV, M4A, MP3, WAV, OGG, Opus, FLAC and AAC.