Aisha AI
by voicelab

AI voice workflow

TTS MP3

The text-to-speech workflow returns audio from text; the documented sync example shows a `.wav` file. If you need MP3, convert the returned audio separately with an audio tool.

Aisha AI · Updated

Try TTS API docs

Use case

Short announcements and audio messages
Reading interface text aloud
Voice prompts for IVR menus
Audio versions of learning materials

Prepare text for a TTS request

Split the script into clear, manageable passages and select the appropriate language for each request. Use the API reference for authentication, supported fields and response details.

Review the returned audio

Open the audio path from the response and listen through the full clip. Check pronunciation, numbers, pauses and sentence endings before saving or passing it to your application.

What to check before launch

Evaluate results with your own material. Keep separate results for each language and recording or playback condition.

Try a representative sample

Generate a short script. Check pronunciation, numbers, pauses and missing words with a fluent listener.

Connect it to the product

Keep API credentials on your server. Check documented formats, limits and response states, and test a failed request before automating the process.

Review each language

Do not generalize one language result to the entire catalog. Ask someone familiar with the target accent and terminology to review the output.

How it works

1

Set the API key in your server environment and prepare the text.

2

Send the text and language to the TTS endpoint documented in the API reference.

3

Retrieve the audio response, listen for errors and save it where your workflow needs it.

Practical guides

How to evaluate multilingual text to speech

Compare TTS voices with a repeatable test for pronunciation, accents, numbers and pacing. Includes sample prompts and a scorecard for each language.

Which workflow fits your use case?

Frequently asked questions

What audio format does a TTS MP3 request return?

The documented sync example returns an `audio_path` ending in `.wav`. Check the downloaded file or media type; if your workflow requires MP3, convert it separately with an audio tool.

How should I review generated speech?

Listen to the returned audio and compare pronunciation, numbers and complete sentence endings against the source text.

Where should the API key be stored?

Keep the key in a server environment variable. Do not expose it in browser code or a public repository.