Up to 4 months free with annual billing on Voibe Dictation and Voibe WorkView pricing
Voibe Logovoibe Resources
Free ToolNo Signup ยท Audio Deleted After

Free Audio to Text Converter and AI Summarizer

Transcribe and summarize meeting recordings, client and sales calls, depositions, interviews and voice memos. Speaker labels and timestamps if you want them. Nobody keeps your audio, and nobody trains on it.

No file handy?

Shapes the AI summary. The transcript is the same either way.

No signup. Your file is transcribed by open-source Whisper on Voibe's zero-retention cloud. The audio is deleted once the transcript is saved and is never used to train any model.

What this converter does

This is a free audio to text converter: upload an audio or video file of up to 30 minutes and get a full transcript, with no account. It accepts MP3, M4A, WAV, MP4, MOV, FLAC, OGG, WEBM and iPhone voice memos. Turn on timestamps and speaker labels for interviews, meetings and subtitles, then copy the text or download it as a TXT file or SRT subtitles.

Every transcript also comes with an AI summary in the format the recording needs. Tell the page what you uploaded and the summary matches it: decisions and action items for a meeting, follow-ups for a client call, objections and next steps for a sales call, topics and witness statements for a deposition, quotes for an interview, or study notes for a lecture.

Transcription runs on open-source Whisper large-v3-turbo on zero-retention providers. Your audio is deleted once the transcript is saved, the transcript is deleted 24 hours later, and nothing is used to train a model. Most free converters ask you to sign up or keep your files in their cloud. This one does neither.

Key Takeaways

QuestionAnswer
CostFree, no signup, no card
File limitUp to 30 minutes and 100 MB per file
FormatsMP3, M4A, WAV, MP4, MOV, FLAC, OGG, WEBM, voice memos
Timestamps and speaker labelsOptional, one switch; needed for SRT
Summary stylesGeneral, meeting notes, client call notes, sales call summary, deposition summary, interview highlights, lecture notes
ExportsCopy, TXT, SRT subtitles (with timestamps on), plus an AI summary
Speech modelWhisper large-v3-turbo (open source, multilingual)
What is keptAudio: deleted once the transcript is saved. Transcript: deleted after 24 hours. Training: never.

How to convert audio to text for free

  1. Drop your file into the box above. Any audio or video file up to 30 minutes works, including an iPhone voice memo exported from the Voice Memos app.
  2. Choose what the recording is. Meeting, client call, sales call, deposition, interview or lecture: the AI summary comes back in that format. The transcript is the same either way.
  3. Tick "Timestamps and speaker labels" if more than one person talks or you want subtitles. Leave it off for voice memos and lectures; the transcript comes back sooner.
  4. Click "Transcribe for free" and keep the tab open. The file uploads over an encrypted connection and transcription starts the moment it lands.
  5. Copy or download the result. Choose TXT for notes and documents, or SRT to add captions to a video in YouTube, Premiere, Final Cut or CapCut.
Animation in three cards. A file called call.m4a slides from your browser into the zero-retention cloud card, where a bar fills as Whisper large-v3-turbo transcribes it. A timestamped, speaker-labelled transcript types into the third card with Copy, TXT and SRT buttons, and the audio file in the cloud card turns red with an Audio deleted stamp, once the transcript is saved. A bar below drains from 24 hours left to deleted for the transcript copy held for the page, and a final mark reads: training on your audio, never, by any model. Closing line: Drop a file, get the text. Keep nothing.
The audio is deleted once the transcript is saved; the transcript follows 24 hours later.

What happens to your file

The file goes from your browser straight to storage over an encrypted connection, without passing through the page you are on. Whisper large-v3-turbo transcribes it on a zero-retention provider, pyannote adds speaker labels when you ask for them, and gpt-oss-120b writes the summary from the text only. The audio is deleted once the transcript is saved. The transcript and summary are kept for 24 hours so this page can fetch them, then deleted. Nothing is used to train any model.

Supported formats and limits

What you haveWorks?Notes
MP3, WAV, M4A, FLAC, OGGYesRead directly
MP4, WEBM videoYesThe audio track is transcribed; SRT export for captions
MOV, AAC, AIFF, other formatsUsuallyConverted first when the audio can be read
iPhone voice memo (.m4a)YesShare from Voice Memos โ†’ Save to Files, then upload
Zoom or Teams recordingYesUpload the audio-only M4A for a smaller file
Files over 30 minutes or 100 MBNoSplit the recording into shorter files
YouTube linksNoUpload a file you own instead

Free audio to text converters compared

Most transcription companies give away some free minutes. The difference is what you trade for them: an account, a monthly cap, or your recordings sitting in their cloud. Free tiers below are from each vendor's own pages, checked October 2026.

ToolFree allowanceWorth knowing
Voibe (this page)Up to 30 min per file, no signupAudio deleted once transcribed, transcript after 24 hours, no training; speaker labels and SRT included
TurboScribe3 files a day, up to 30 min eachUnlimited plan is $10/month billed annually ($120/year)
Riverside free transcriber"Unlimited transcriptions for free. No sign up required"Free tier assigns all text to one speaker; TXT or SRT
Otter free plan3 file imports per accountAccount required; privacy policy says it trains on de-identified audio and transcriptions
Rev free plan45 AI minutesAccount required; Essentials is $29.99/month

For a deeper look at paid options, see our TurboScribe alternatives guide and the best transcription apps for Mac. For live meetings, compare the best meeting transcription apps.

Transcribe and summarize recordings by use case

The transcript is the same whatever you upload. What changes is the summary: pick the matching option under "What is the recording?" and it pulls out what that kind of recording is for. Here is what to choose for each.

RecordingChooseSpeaker labelsThe summary pulls out
Meeting recordingMeeting notes and action itemsOnOverview, decisions, action items with owners and dates, open questions
Client callClient call notesOnClient goals and concerns, what was agreed, follow-ups for each side, figures and deadlines
Sales callSales call summaryOnNeeds and pain points, objections and answers, budget, competitors, next steps
DepositionDeposition summaryOnTopics in order, key statements and admissions, objections, exhibits
Interview or research callInterview highlightsOnMain themes, word-for-word quotes, facts, figures and names
Lecture or classLecture or class notesOffKey concepts, definitions, examples, assignments and dates
Voice memo or podcastGeneral summaryOff (on for podcasts)The main points in a few lines

Meeting recordings: transcript, minutes and action items

Upload a Zoom, Microsoft Teams or Google Meet recording and choose "Meeting notes and action items". You get the full transcript with who said what, plus meeting minutes: a one-line overview, the decisions made, action items with owners and due dates when they were said, and open questions. Our Zoom recording guide shows where Zoom saves the audio file. If you take meeting notes every week, Voibe Work captures the call from your computer with no bot joining and no file to upload.

Client calls: call notes and follow-ups

For advisors, consultants, agencies and lawyers, choose "Client call notes". The summary lists the client's goals and concerns, what was agreed, the follow-ups each side owns and any figures or deadlines mentioned, ready to paste into your CRM or a follow-up email. Get everyone's agreement before you record: you are responsible for recording-consent rules where you and your client are.

Sales calls: objections, budget and next steps

Upload a discovery or demo call and choose "Sales call summary". It pulls out the prospect's needs and pain points, every objection and how it was answered, any budget or pricing discussed, competitors named, and the agreed next steps with dates. With speaker labels on, the transcript separates the rep from the prospect, which makes it easy to review a call or coach on it.

Depositions: transcript and deposition summary

Upload the deposition audio or video, turn on speaker labels and choose "Deposition summary". You get a timestamped transcript plus a neutral summary: the topics covered in the order they came up, key statements and admissions by the witness, objections noted and exhibits mentioned, with no legal conclusions drawn. This is an AI transcript, not a certified court reporter's transcript, so use it to review testimony and prepare, and check quotes against the recording. More tools for legal work are in our legal transcription guide.

Interviews and research calls: quotes you can check

Choose "Interview highlights" for the main themes, word-for-word quotes and the facts, figures and names mentioned. Every transcript line carries a timestamp, so you can find a quote and check it against the recording before you publish or cite it.

Lectures, podcasts and voice memos

Choose "Lecture or class notes" for study notes with key concepts, definitions, examples and any assignments or dates. For a voice memo, leave it on "General summary": a rambling five-minute memo comes back as the few points you meant to make. For a podcast, turn on timestamps and download the SRT to publish captions.

When to use something other than this converter

This page is built for one file at a time. If you transcribe meetings every week, use Voibe Work, which records Zoom, Teams and Google Meet calls from your Mac or Windows PC with no bot in the call and saves notes on your computer. If the recording must never leave your computer, use an on-device tool: Voibe dictation runs Whisper locally on Apple Silicon Macs, and MacWhisper transcribes files locally. If you want to write by speaking rather than transcribe a file, use Voibe dictation in any app.

Frequently Asked Questions

Use Cases

How do I summarize a meeting recording?
Upload the recording, choose "Meeting notes and action items" under "What is the recording?", and tick "Timestamps and speaker labels" so you can see who said what. You get the full transcript plus notes with an overview, decisions, action items with owners and dates when they were said, and open questions. Zoom, Teams and Google Meet recordings all work; the audio-only M4A file is the smallest to upload.
Can I transcribe and summarize a sales call?
Yes. Upload the call recording and choose "Sales call summary". The summary pulls out the prospect's needs and pain points, objections and how they were answered, budget or pricing mentioned, competitors named, and the agreed next steps. Turn on speaker labels to separate the rep from the prospect in the transcript.
Can I transcribe a client call?
Yes. Choose "Client call notes" for a summary of the client's goals and concerns, what was agreed, follow-ups for each side and any figures or deadlines mentioned. Make sure everyone on the call agreed to the recording first: you are responsible for recording-consent rules where you and the client are.
How do I make a deposition summary from a recording?
Upload the deposition audio or video, tick "Timestamps and speaker labels", and choose "Deposition summary". You get a timestamped, speaker-labelled transcript plus a neutral summary of the topics covered in order, key statements and admissions by the witness, objections noted and exhibits mentioned. It is an AI transcript, not a certified court reporter's transcript, so use it to review and prepare, and check quotes against the recording before relying on them.
Do I need consent to record and transcribe a call?
Often, yes. Recording-consent laws differ by place: some US states and many countries require everyone on the call to agree before it is recorded. Get consent before you record. This page transcribes the file you upload and does not check how it was recorded.

Using the Converter

Is this audio to text converter really free?
Yes. There is no signup, no card and no watermark. Each file can be up to 30 minutes long and 100 MB, and there is an hourly limit per visitor so the free minutes stay available for everyone.
Which file formats can I upload?
MP3, WAV, M4A, MP4, FLAC, OGG and WEBM go straight in. Other audio and video files, such as MOV, AAC, AIFF or iPhone voice memos, are converted first when their audio can be read. The format is detected from the file's content, not its extension.
Can it tell who is speaking?
Yes. Tick "Timestamps and speaker labels" before you start and the transcript marks each turn as Speaker 1, Speaker 2 and so on, with a timestamp on every line. Speaker labelling is the slow part of transcription, so leave it off for single-speaker recordings like voice memos or lectures.
How long does a transcription take?
There are two parts: a wait in the queue until a transcription worker picks up your file, then the transcription itself, which takes a fraction of the recording's length. For short clips the queue is most of the wait. Speaker labels add a little time. Keep the tab open; the page shows which stage you are in, and the transcript appears when it is ready.
What languages does it support?
The speech model is Whisper large-v3-turbo, which is multilingual, so the recording does not have to be in English. The summary is written by a multilingual model too.

Privacy and Data

What happens to my audio after I upload it?
It is encrypted in transit, transcribed by open-source Whisper on Voibe's zero-retention cloud, and deleted once the transcript is saved; anything left over is removed within a day. The transcript and summary are kept for 24 hours so this page can show them to you, then deleted automatically. Nothing you upload is used to train any model.
Does OpenAI, Google or another AI lab see my file?
No. Transcription runs on Whisper large-v3-turbo, speaker labels on pyannote, and the summary on gpt-oss-120b, all open-source or open-weight models running on zero-retention providers. No third-party AI lab account is involved and you never paste an API key.
Should I upload confidential recordings?
For recordings that must never leave your computer at all, use an on-device tool instead. Voibe's dictation app can run Whisper entirely on Apple Silicon Macs, and MacWhisper transcribes files locally. This converter is a cloud tool: zero retention limits what is kept, but the file does travel to a server to be transcribed.

Exports and Next Steps

Can I get subtitles (SRT) from a video?
Yes. Tick "Timestamps and speaker labels", upload the MP4, MOV or WEBM file, and click "Download .srt subtitles" when the transcript is ready. Without that switch the transcript comes back as plain text with no timings, so there is no SRT. Each subtitle line carries the start and end time from the transcript, and speaker names are included when speaker labels are on.
Can Voibe transcribe my meetings without a recording file?
Yes, with Voibe Work. It captures Zoom, Teams, Google Meet and other calls from your Mac or Windows PC with no bot in the call, and saves the transcript and summary on your computer. It reminds you to tell everyone you are taking notes when recording starts.

Sources and References

  1. Voibe transcription documentation (models, formats, retention: audio deleted once the transcript is saved; transcripts deleted 24 hours after the job finishes)
  2. Voibe cloud AI privacy (zero-retention providers, open-source models)
  3. Riverside free audio transcriber (free tier terms, retrieved 2026-10-28)
  4. Otter privacy policy (training on de-identified audio recordings and transcriptions)
  5. Rev pricing and TurboScribe pricing

Got a recording to transcribe?

Drop it into the converter at the top of this page and get the text back, with speaker labels and timestamps if you want them.

  • Free, no signup
  • MP3, M4A, WAV, MP4 and more
  • Audio deleted once transcribed
  • Never used for training
Transcribe a file now