ZVUKOGRAM

What do tokens charge for?

The offers page lists a free token option and paid token packages; each voice-over deducts tokens, and the homepage equates one token with one ruble.

• Free option — 10 tokens — offers page, checked 2026-10-31.
• 150-ruble option — 150 tokens — offers page, checked 2026-10-31.
• 250-ruble option — 262.5 tokens — offers page, checked 2026-10-31.
• 500-ruble option — 550 tokens — offers page, checked 2026-10-31.
• 1000-ruble option — 1150 tokens — offers page, checked 2026-10-31.
• 3000-ruble option — 3600 tokens — offers page, checked 2026-10-31.

The pricing page says ordinary-voice speech synthesis costs from 1 token per 1000 characters, premium-voice synthesis from 5 tokens per 1000 characters, and transcription from 2.5 tokens per minute. The offers page says premium voices cost more because their neural networks consume substantially more resources for speech generation.

The vendor’s FAQ says charges are based on characters and that repeating the same text with the same voice is free. Check how that repeat-generation rule applies to the intended workflow before estimating token use.

Which files and voice tools are listed?

Subtitle voice-over accepts SRT, VTT, and SUB files and produces an audio track with timings, according to the homepage. The transcription tool recognizes speech from audio and video and provides the text in Word.

The FAQ says the editor accepts text, DOCX, PDF, and TXT, and can create one audio file with dialogue spoken by two voices. It also lists SSML tags for speech and pronunciation control, plus a user-editable pronunciation dictionary that applies across generations when enabled.

The homepage lists over 54,000 sound effects in MP3, WAV, and OGG, and over 10,000 AI music tracks for videos and podcasts. The FAQ says users can add library music and sound effects, download audio in MP3, OGG, OPUS, WAV, M4A, WEBM, and FLAC formats, save generated audio, or share it through a player-page link.

What should support clarify before registering?

Ask how the commercial-use statement applies to the specific audio you plan to publish, and which document governs that use. In general, a buyer should get the permitted channels and any attribution conditions stated in terms they can retain.

Clarify how token deductions work for the intended mix of ordinary and premium voices, including repeated generations. In general, metering questions matter most when a workflow relies on revisions or large batches.

Ask whether your planned use requires rights beyond the service’s terms, such as rights in source material or background audio. In general, the buyer remains responsible for checking rights attached to each asset.

How do the API requests return audio?

The API sends HTTP requests for neural text-to-speech and returns audio files; the provider says access requires adding funds to the account first. Every request must include the account’s secret token and registered email address.

For immediate output, the provider documents the /text method, which returns an audio file and supports texts up to 1,000 characters per request. For longer jobs, /longtext supports texts up to 1,000,000 characters and processes them asynchronously; /result is used after /longtext or /subs, not after /text. The /subs method returns timing information linking text fragments to audio time.

The API accepts form-encoded POST, JSON POST, and GET query-string requests, and /text supports MP3, WAV, OGG, OPUS, and FLAC output. Optional settings include speed, pitch, style, role, pauses, volume, and audio effects; background music can be added with volume and looping controls. The provider also documents project-and-audio deletion, sample code in PHP, Python, cURL, and JavaScript, and integrations with n8n, Make, and Zapier.

Which stated terms add working options?

• The homepage lists over 140 Russian voices and over 3,000 voices in 150 languages, giving buyers voice and language options to check against their production needs.
• Subtitle voice-over keeps timings in the produced audio track, according to the homepage.
• The API separates immediate narration, asynchronous long-text processing, and timed-fragment output into documented methods.
• The API includes project deletion, while the FAQ describes a pronunciation dictionary that can apply across audio generations when enabled.
• The provider lists integrations with n8n, Make, and Zapier, alongside developer support and updates through its Telegram community channel.

Which rules narrow audio use?

• The DMCA page says ZVUKOGRAM does not provide licenses or determine whether audio is free to use.
• The site prohibits public playback and online streaming of identified copyrighted audio from its pages.
• ZVUKOGRAM says users bear the risks of using or publishing downloaded audio on their own or other resources, including for commercial use. The DMCA page also says music and sounds used on YouTube may require additional rights or attribution in the description.
• The API documentation says each request must include the account’s secret token and registered email address.

The unresolved purchasing details are listed under What we could not confirm.

Which voice-service terms should match?

Compare voice and language coverage, supported file formats, and limits on text or audio length. In general, check whether rights terms cover the specific publication channels and source material involved. Also compare how usage is metered, how API requests return results, what trial access includes, and what support covers.

Who can use the listed API methods?

The provider presents the API for automated narration, TTS integrations, bulk text narration, and audio content generation. A workflow that needs immediate audio can use the documented /text method, while long-text processing and timed text-to-audio fragments have separate methods.

The provider also lists integrations with n8n, Make, and Zapier. API access requires funds in the account before use, so check that condition against the planned workflow.

What do the API and offer pages answer?

Is repeating a voice-over charged again?

The vendor’s FAQ says repeating the same text with the same voice is free. It also says charges are based on characters, so the stated repeat rule applies to that same-text, same-voice condition rather than establishing a general exemption for revisions.

Which API methods return audio directly?

The API documentation says /text returns an audio file immediately. The provider describes /longtext as asynchronous and /subs as returning timing information that links text fragments to audio time; /result is used after the latter methods, not /text.

Which audio formats does the FAQ list?

The FAQ lists MP3, OGG, OPUS, WAV, M4A, WEBM, and FLAC among the formats available for audio downloads. This is the FAQ’s stated list, not a guarantee that every format is supported by every API method.

What does the site say about commercial use?

The homepage says a commercial license is included in the plan, while the DMCA page says the site does not provide official licenses or determine whether audio is free to use. The same DMCA page assigns users the risks of using or publishing downloaded audio.

What does the homepage say about Russian cards?

The homepage says the service works in Russia without a VPN and accepts Russian cards. This describes the provider’s own statement about access and payment; it does not resolve the separate licensing terms for particular audio.

Where does the provider direct API developers?

The provider lists developer support and updates through its Telegram community channel. Its API documentation also describes integration with n8n, Make, and Zapier.

Are the token and rights terms workable?

Fits if the workflow can use the stated token charges and the buyer has clear rights for each audio asset.

Probably not if the buyer needs official licensing documents, which the DMCA page says the provider does not supply.

Main risk… the provider’s pages give different signals about licensing, while the DMCA page places use and publication risks on the user.

Sources

Below are the public pages behind the facts above, together with the date each one was read.

• zvukogram.comchecked 2026-10-31
• zvukogram.com/index.php?r=site/offerschecked 2026-10-31
• zvukogram.com/node/faqchecked 2026-10-31
• zvukogram.com/node/apichecked 2026-10-31
• zvukogram.com/node/dmcachecked 2026-10-31

Notes on the making of this review

Editorial team of Profile Workshop. Profile Workshop compiled this page from the public sources above as they stood on 2026-10-31. We did not test the service ourselves. We revise the text whenever those sources are updated.

Last updated: 2026-10-31

Right of reply

Do you represent ZVUKOGRAM and see an error or outdated information here? Write to us with evidence. Once the correction is documented, we fix the entry or take it down. Payment never gets an entry removed or softened.

Our contact for replies

No data available

Response time

No data available

Contact · Affiliate disclosure

More write-ups filed under Text to speech

• Voicegenerator.io — Text to speech
• Naturalreaders — Text to speech