Sonicribe Cloud Language Support: How to Evaluate Your Languages

Language support depends on the selected service

Choose the language you speak in Settings → Listening → Speech language. For mixed-language dictation, use Auto-detect. Record a short passage with the names and specialist terms you use, then choose the speech model and word groups that fit your writing.

Transcription, automatic language detection, mixed-language recognition, and translation are different capabilities. Support for one does not prove support for the others. This guide gives you a practical way to evaluate them without assuming a blanket language count or accuracy guarantee.

Sonicribe requires an Apple Silicon Mac, Google sign-in, and internet. Windows and Intel Macs are unsupported. It sends audio to online services; optional formatting sends text for online processing too.

Define what you need

Start with a short list of your actual requirements. Include the spoken language, regional variety, output script, and any secondary language you mix into conversation.

For example, someone dictating a Korean project update might also say an English product name and a numeric release date. An English-only sample would not test that workflow. Someone preparing a bilingual interview needs to distinguish transcription in the original language from a separate translation.

Use this worksheet before evaluating:

Record a small test set

Prepare three non-sensitive samples: everyday speech, a passage with names and numbers, and a passage mixing languages if you use that pattern. Keep a written reference so you can identify errors rather than judging fluency alone.

Select an available language setting when your chosen service offers one. If detection is available, compare a detected-language result with your intended language. Check the output script as well as the words.

Review names, dates, currencies, units, and negative statements carefully. Repeat the same sample if you change service or language settings so that you can compare results fairly.

Evaluate translation separately

A transcript should represent what was spoken. A translation changes the output language. Optional formatting can change presentation again. Keep these steps separate during evaluation so you can tell where a mistake entered.

Do not assume Sonicribe produces both an original transcript and a translation, or translates every language supported by a speech service. Confirm the current workflow before relying on it.

If a translation will be published or used for an important decision, have someone proficient in both languages review it. Check idioms, names, units, and tone against the source.

Test optional formatting with the original beside it

If you use online formatting, compare the formatted version with the transcript. It should preserve the content you need, including language-specific terms and mixed-language names. A polished paragraph is not proof of a faithful result.

Where vocabulary settings are available, test a few terms before making a larger list. A vocabulary entry does not guarantee recognition, and behavior need not be identical across languages or services.

Plan credits and connectivity

Eligible new accounts receive 100 free minutes with Qwen3-ASR, with no time limit. Pro costs US$7/month for up to about 2,500 minutes a month; Max costs US$20/month for up to about 10,700 minutes a month with Qwen3-ASR. Speech and optional formatting share the balance, unused paid credits reset monthly, and recording pauses at zero credits.

Measure the credit change during each test. No fixed words-per-language allowance is established here. Sonicribe cannot provide offline field transcription; plan a separate local workflow if connectivity is unavailable.

Use non-sensitive interview or workplace examples to build your language setup. Keep a short sample with names and numbers, and read the privacy policy for the online processing details.

Decide from the results

Choose the service and settings that work for your actual language mix, then repeat a sample when you change them. Keep the written reference and correction notes so you can recognize a change in output quality.

See pricing, the Cloud trial guide, and the multilingual transcription guide for related planning.