Text to speech
ZVUKOGRAM
ZVUKOGRAM charges tokens for text-to-speech and transcription, while its homepage says a commercial license is included. Its DMCA page, however, says copyrighted materials are for private listening and that the site does not provide official licenses, leaving buyers to clarify which audio the commercial-use wording covers. This matters to media buyers choosing between generated narration and audio from the site’s library.
Editorial review based on public sources, without hands-on testing; facts checked on 2026-10-31.
This page contains no partner link to this service.
- Logo
Logos are trademarks of their respective owners and shown only to identify the service.- Name
- ZVUKOGRAM
- Category
- Text to speech
- Official site
- zvukogram.com · No link provided
- Facts checked
Check the listed audio tools
The provider’s homepage, checked 2026-10-31, lists these tools and counts.
- Audio tools: speech synthesis, subtitle voice-over, transcription, sound effects, and AI music.
- Speech voices: over 140 Russian voices and over 3,000 voices in 150 languages, according to the homepage.
- Sound library: over 54,000 effects in MP3, WAV, and OGG formats, according to the homepage.
- AI music: over 10,000 tracks for videos and podcasts, according to the homepage.
- Subtitle narration: accepts SRT, VTT, and SUB files and produces audio with timings.
- Transcription: recognizes speech from audio and video and provides text in Word.
Compare token charges by task
The pricing page treats tokens as the billing unit: each voice-over deducts tokens, and the provider says one token equals one ruble. Its offers page lists a free-token option and paid token options, each with its own stated amount.
- Free option: 10 tokens, according to the offers page checked 2026-10-31.
- 150-ruble option: 150 tokens, according to the offers page checked 2026-10-31.
- 250-ruble option: 262.5 tokens, according to the offers page checked 2026-10-31.
- 500-ruble option: 550 tokens, according to the offers page checked 2026-10-31.
- 1000-ruble option: 1150 tokens, according to the offers page checked 2026-10-31.
- 3000-ruble option: 3600 tokens, according to the offers page checked 2026-10-31.
For ordinary-voice text-to-speech, the offers page lists a rate from 1 token per 1000 characters; premium voices start at 5 tokens per 1000 characters. The same page lists transcription from 2.5 tokens per minute, so compare the task-specific meter rather than treating every audio operation as a character charge. The provider says premium voices cost more because their neural networks consume substantially more resources for speech generation.
Match inputs to voice outputs
The FAQ describes an editor that turns text into speech using neural voices and accepts text, DOCX, PDF, and TXT input. It also supports a single audio file with dialogue spoken by two voices.
For speech control, the editor offers SSML tags and a user-editable pronunciation dictionary that applies across generations when enabled. Background music and sound effects from the library can accompany the narration.
The FAQ lists downloads in MP3, OGG, OPUS, WAV, M4A, WEBM, and FLAC, among other formats, and says generated audio can be saved and shared through a player-page link. Check that the output and sharing options fit the intended delivery workflow.
Put rights questions to support
Before relying on the included commercial-license wording, ask what it covers for generated voice output versus audio drawn from the site’s library. The homepage says the commercial license is included, while the DMCA page says copyrighted materials are for private listening and further use is not allowed.
Ask which rights apply to a specific file and whether the restriction concerns the site’s identified copyrighted audio or other material. The DMCA page also says the site does not determine whether audio is free to use and does not provide official licenses or other use documents.
The provider says its audio files come from open or described-as-free sources, but also places risks of using or publishing downloaded audio on users. Request a file-specific explanation of the applicable terms before using library audio in a commercial project.
Trace text through the API
The API documentation describes two text routes: `/text` returns audio immediately, while `/longtext` processes text asynchronously. The documented limits are 1,000 characters per `/text` request and 1,000,000 characters for `/longtext`. The `/subs` method returns timing information that links text fragments to audio time.
The documentation accepts form-encoded POST, JSON POST, and GET query-string requests; `/text` output can be MP3, WAV, OGG, OPUS, or FLAC. Optional speech settings cover speed, pitch, style, role, pauses, volume, and audio effects. Background music can be added, with volume and looping adjustable.
API access requires funds in the account first, and each request must include the account’s secret token and registered email address. The `/result` endpoint applies after `/longtext` or `/subs`, not after `/text`. The documentation also provides sample code in PHP, Python, cURL, and JavaScript, lists integrations with n8n, Make, and Zapier, and includes an endpoint to delete a project and its audio file from the server.
Account for the stated restrictions
The API requires funds before access, and each request needs the account secret token and registered email address. Its `/result` endpoint is reserved for asynchronous `/longtext` or `/subs` work, rather than immediate `/text` requests.
The copyright terms complicate reuse of site audio: the provider limits identified copyrighted materials to private listening, prohibits public playback and online streaming of that audio, and says it supplies no official use documents. The homepage’s included commercial-license statement therefore needs clarification against the DMCA page’s limits.
For the same text generated again with the same voice, the FAQ says no repeat charge applies; the provider otherwise describes token deductions for characters. The provider also says users bear risks tied to downloaded audio, including commercial use, and that YouTube use of music and sounds may require additional rights or attribution. Several details are missing; they are listed under What we could not confirm.
Use the documented workflow options
The provider’s API documentation and FAQ describe workflow options that can be checked against the task.
- Subtitle narration accepts timed subtitle files and returns audio with timings, a documented option when the source material is already subtitled.
- The editor accepts text and document inputs, supports dialogue between two voices, and offers SSML and an editable pronunciation dictionary.
- API users can choose immediate or asynchronous text processing, or request timing information linking text fragments to audio time.
- The API lists speech controls, background-music adjustments, and a project-deletion endpoint. It also names n8n, Make, and Zapier integrations.
Compare voice tools by terms
Compare providers by voice and language coverage, supported file formats, text limits, commercial-use wording, billing meter, API access, trial terms, and support. Ask each provider to clarify whether the license covers generated narration, library audio, or both, and check how its stated restrictions apply to your intended use.
Match token use to the task
The documented scope covers text-to-speech, subtitle narration, transcription, and API-based narration. The provider charges tokens for speech generation and transcription, with separate meters for ordinary voices, premium voices, and transcription.
For a planned workflow, estimate the text volume or audio duration and apply the relevant rate from the provider’s offers page. The feature list does not resolve whether the included commercial license covers the audio you intend to use, so check that scope before relying on it.
What we could not confirm
- Voice samples, voice-level language coverage, and independent output-quality evidence.
- Account registration, payment methods, refunds, and token-expiry conditions.
- Support response times and channels.
Resolve five specific term checks
How do ordinary and premium voice charges differ?
The provider lists ordinary-voice text-to-speech from 1 token per 1000 characters and premium-voice text-to-speech from 5 tokens per 1000 characters. Those are separate starting rates, so identify the voice category before estimating token use.
How is transcription charged?
The provider lists audio transcription from 2.5 tokens per minute. This is a duration-based meter, unlike the character-based rates it gives for voice generation, so estimate transcription separately from narration.
When does the API use the result endpoint?
The API documentation says `/result` applies after `/longtext` or `/subs`, not after `/text`. The `/text` method returns audio immediately, while the other listed methods process text asynchronously or return timing information.
Does repeating a generation incur another charge?
The vendor says tokens are charged for characters, while repeating the same text with the same voice is free. The exception requires both the text and voice to match, so ask how changed text or a different voice is billed.
What does the commercial license cover?
The homepage says a commercial license is included, but the DMCA page limits copyrighted materials to private listening and says further use is not allowed. The provider also says it does not provide licenses or determine whether audio is free to use.
Does the license settle YouTube use of music and sounds?
No blanket clearance follows from the stated license wording. The DMCA page says YouTube use of music and sounds may require additional rights or attribution in the description, so check the terms for the specific audio file.
Proceed only with rights clarity
Fits if the intended task is text-to-speech, subtitle narration, transcription, or API-based narration, and the applicable audio rights are clear.
Probably not if the workflow depends on using copyrighted library audio commercially without clarification; the DMCA page states a private-listening restriction.
Main risk… the commercial-license wording does not explain how it interacts with that restriction. Resolve the rights scope first, then match the API method and token meter to the task.
Sources
These public pages are the basis of the facts above; next to each is the date it was read.
- zvukogram.com/index.php?r=site/offerschecked
- zvukogram.comchecked
- zvukogram.com/node/dmcachecked
- zvukogram.com/node/faqchecked
- zvukogram.com/node/apichecked
How this page was prepared
Editorial team of Funnel Notebook. Source-based work: each fact comes from a page listed above, read on 2026-10-31 by the Funnel Notebook editors. There was no hands-on testing. The page is updated when the sources change.
Last updated:
Right of reply
Speaking for ZVUKOGRAM? If you believe the entry is inaccurate or stale, send us proof. We amend or withdraw the entry as soon as the refutation is on record. Money does not remove or soften an entry here.
- Our contact for replies
- No data available
- Response time
- No data available