Questions and answers

Yes, a free one: sign up with your email or with Google. The sign-in page also has a demo with sample data, so you can look around the app before you sign up.

Confirming your email adds 150 credits: 15 minutes of transcription or live subtitles, or about 3,000 characters of speech with Standard voices. With Enhanced and Premium voices, free credits voice clips of up to 300 characters each, 600 in total, until your first purchase. Voice previews are always free, and no card is needed.

No. Speecho runs in the browser on computers, tablets and phones. There is no app to download and no bot joining your meetings: you upload a recording or start a live session yourself.

Recent Chrome, Edge, Safari and Firefox all handle uploads, text to speech and live sessions from the microphone. Taking the sound of another browser tab (a call or a video playing in the browser) works only in Chrome and Edge on a computer: Safari does not hand tab audio to websites, and Firefox shares the picture only.

15: English, Spanish, German, French, Portuguese, Italian, Dutch, Polish, Vietnamese, Turkish, Ukrainian, Arabic, Japanese, Korean and Chinese (Simplified). Switch with the language menu at the top of the app.

Everything is paid from one balance of credits. One credit is 6 seconds of transcription, so 10 credits are a minute and 600 an hour. Packs: $5 = 2,400 credits, $10 = 6,000, $25 = 16,800 (1,800 of them bonus), $50 = 36,000 (6,000 bonus). That is $0.83 to $1.25 per hour of transcription, depending on the pack. Details on the pricing page.

• Transcription and live subtitles: 10 credits a minute, billed by the second.
• Conversation: 10 credits a minute, 40 with the translation read aloud.
• Interpreter: 40 credits a minute on Standard, 80 on Premium.
• Text to speech: 50, 110 or 160 credits per 1,000 characters on Standard, Enhanced or Premium voices, at least 10 per clip.
• Burned-in video captions: 30 credits per video minute, transcription included.
• Handwriting scan: 40 credits a page.
• Asking AI about a transcript: from 5 credits a question.

Chapters, the follow-up email draft and merging transcripts are free, and so are the first two summary styles and the first two translations of every transcript. The app shows the price, or how long your balance lasts, before you start. The full list is on the pricing page.

No. Credits never expire, and there is no monthly reset. Top up once and use them whenever you need.

No. You buy a pack when you need credits. Nothing renews, and your card is never charged again on its own.

The discount comes as bonus credits: the $25 pack includes 1,800 and the $50 pack 6,000. The $50 Studio pack can be bought up to 5 at once ("Buy 2 or more" in the top-up window), and every pack after the first adds 2,000 more: two packs give 74,000 credits for $100, five give 188,000 for $250.

Paddle, our Merchant of Record, runs the checkout and issues the receipt and the invoice. Prices are in US dollars, and VAT may be added at checkout depending on your country; a VAT-registered business can enter its VAT ID there. There are no seat fees, setup fees or minimums.

Paddle shows the payment methods available in your country at checkout. Card details go to Paddle and never reach Speecho.

Yes. Within 14 days of a purchase you get a full refund for any reason, even if you used the credits. After 14 days, refunds cover failures on our side. Email support@speecho.app with the order number from your Paddle receipt; the money goes back to the original payment method. The full policy is on the refund page.

No. When a transcription, a scan or a video render fails, its credits go back to your balance on their own, and the line in your history says whether to try again in a few minutes or whether the file itself could not be read. If a file finishes but the text is empty or useless (music drowning out faint speech, for example), write to us and we return the credits.

Audio: MP3, M4A, WAV, WebM, OGG, AAC, FLAC, AIFF and Opus, up to 200 MB and 5 hours per file. Video: MP4, MOV and WebM, up to 4 hours. Your browser takes the sound out of a video before the upload, so only the audio travels and the video stays on your computer.

Both, from the same account. Upload a recording and the transcript comes back in minutes, or start live subtitles and the text appears while people speak, also on a link others open on their phones.

Usually a few minutes for an hour of audio. You can close the tab meanwhile: the finished transcript is also sent to you by email.

On clean audio 90 to 97% of words come out right. Noise, people talking over each other and a distant microphone lower it. Setting the language and adding names and terms to the glossary (Names and terms) before you upload raise it. The glossary stays for every later transcription and also works in live subtitles.

Yes. Sentences get their punctuation, and the text is split into paragraphs with timestamps, so it reads without a pass by hand. More on the speech-to-text page.

Uploaded recordings: 99 languages, detected automatically. A finished transcript translates into 18 languages. Live subtitles recognise 60 languages and can translate into any of them while people speak.

Yes. Transcribe each file, select them in your history and press Merge. The parts come together in the order you set, with timestamps running on. Merging is free, and the originals stay.

Yes. Click the pencil next to any paragraph, fix a name or a term and save. Every download uses the corrected text.

Plain text, Markdown, Word (.docx), PDF, and SRT or VTT subtitles.

Write a summary in the style you need (meeting minutes, interview, lecture, podcast notes and more), split the recording into chapters with timestamps, draft a follow-up email, answer your questions about the recording, translate it and remove filler words. Answers come from the transcript only. What each one costs is under What does each feature cost?

Live subtitles write down what is said, and can translate it, on your screen and on a link other people open on their phones. Conversation is for two people with two languages and one phone: each reads the other on their half of the screen, and the translation can be read aloud. Interpreter speaks the translation in a voice, for listeners who would rather hear it than read it.

No. They open a link, usually from a QR code, in their phone's browser. Nothing to install, no sign-up, nothing to pay: the session is billed to you. Up to 20 people can listen to one interpreter session.

Yes, and nothing joins the call as a bot. Open the call in a Chrome or Edge tab and pick that tab as the audio source. Live subtitles can also go into Zoom or YouTube Live as the platform's own captions (Captions in a meeting); Google Meet does not accept outside captions, so there you share the viewer link instead.

Standard reads the translation in one voice you pick, runs about five seconds behind the speaker, follows your glossary and covers more than 50 languages. Premium uses several voices, runs about three seconds behind and covers more than 70. Standard costs 40 credits a minute, Premium 80; you choose before each session.

Live subtitles and Conversation: pick 1, 2 or 4 hours before you start (2 by default). During a session you can add an hour at a time, up to 4 hours, as far as your balance covers, and a session that reached its limit can be continued from your history. Silence does not stop these sessions, so an open tab keeps running until you stop it or its time runs out. An interpreter session lasts up to 2 hours.

Short outages are fine. Speecho keeps listening, holds on to what is said while the connection is down and reconnects on its own, and the offline time is not billed. If the connection stays down for about a minute, the session ends and what was said is saved to your history.

Yes. When you stop, the text and its translation are saved to your history like any transcript, ready to download as text or subtitles, summarize or share. The audio is not kept unless you turn on Keep audio before you start, and a translated voice is never stored.

Yes. Add them to the glossary before the session, and set how a term should be translated if it has a fixed translation of its own.

Standard voices (50 credits per 1,000 characters) are clear and good for everyday narration. Enhanced (110) and Premium (160) use ElevenLabs voices that sound more natural and expressive. Every voice has a free preview. More on the text-to-speech page.

Standard reads 63 languages; Enhanced and Premium read 87, with native voices in 31 of them. Paste your text and the app picks up its language and suggests voices for it. The voice library has a sample of each.

Yes, with tags in the text: [pause] and [long pause] add a pause, and tags such as [whispering], [excited] or [laughs] change the delivery. How closely a voice follows a tag depends on the voice, so try a sentence first. More on the text-to-speech page.

Yes. Switch Text to speech to Dialogue: each character gets a voice, up to 8 characters and 100 lines, and you download one MP3. A dialogue costs the same as the same text in one voice. More about dialogues.

Yes. Switch to List and paste up to 1,000 lines, or open a .txt or .csv file: each line becomes its own MP3, named after the line, and the set downloads as one ZIP. More about lists.

Yes. Switch to Subtitles and open an SRT, VTT or SBV file: every subtitle is voiced at its own time, as one MP3 track and as separate files in a ZIP. More about voice-overs from subtitles.

Yes. Make a podcast from text turns an article, a chapter or your notes into a conversation between two hosts of about 5, 10 or 20 minutes, and you can check the script in the Dialogue editor before it is voiced. More about podcasts from text.

Yes. In a long text voiced in parts, or in a dialogue, you can voice any part or line again on its own and pay only for that part.

A document or a book of up to 250,000 characters, about four hours of speech, is voiced in one pass, in parts you download joined into one MP3 or, for a book, as chapter files and an M4B audiobook. A longer book is voiced in several passes.

An MP3 you can play in the browser and download; Enhanced and Premium clips come with SRT or VTT subtitles at no extra cost. Clips are kept for 90 days, or less if you choose a shorter period, and you can delete any of them sooner.

Yes, the generated audio can be used commercially; you are responsible for the text you voice. Speecho does not offer voice cloning.

Not by default. Uploaded and live audio is used to make the transcript and then deleted. If you turn on Keep audio before an upload or a live session, a compressed copy stays next to the transcript so you can play it back; it is deleted with the transcript, or sooner if you delete it yourself.

90 days, after which they are deleted automatically together with kept recordings and voice clips. On the account page you can choose 30 or 7 days instead (Keep my transcripts for), and you can delete anything sooner.

No. The AI providers process your audio and text only to return the result, and none of them train models on it.

The providers listed by purpose (speech recognition, translation, voices, text) in the privacy policy. Payments go through Paddle. How the service is secured is on the security page.

Most of the free tools work entirely in your browser, and the file never leaves your device. Each tool's page says so, and says when something does leave it, for example a browser voice that reads text through Google or Microsoft.

On the account page in the app. Download all my data gives a zip with your transcripts, payments and account details. Delete account removes the account, the balance and all history at once, and cannot be undone.

With your email and password, or with Google. On the account page you can add a passkey (fingerprint, face or device PIN) and turn on two-factor sign-in with an authenticator app.

The status page shows whether the app and its services are up. It is hosted separately, so it answers even when Speecho does not. If everything there is green, write to us with your browser, your device and what happened.

Email support@speecho.app. The person who builds Speecho answers, usually within a business day and always within five. Include the email of your account and, for billing, the order number from your Paddle receipt. More on the contact page.