# What is the difference between text to speech and speech to text?

Text to speech turns written words into a voice. Speech to text turns a voice into written words. See what each one does, where it is used, and what it costs.

Published 2026-11-06. Paxa Labs.

The two names hold the same words in opposite order, and they describe opposite jobs. Text to speech reads written text aloud in a synthetic voice. Speech to text listens to a recording or a live voice and writes down what was said.

## How do text to speech and speech to text differ?

|  | Text to speech | Speech to text |
| --- | --- | --- |
| Also called | TTS, speech synthesis, AI voice generation | STT, transcription, speech recognition, ASR |
| Takes | Written text | Audio of someone speaking |
| Returns | An audio file or a live audio stream | A transcript, a caption file, or live text turn by turn |
| Priced by | Characters of text | Minutes of audio |
| Paxa model | Paxa TTS Flash v2 | Paxa STT Lite and Paxa STT Lite Realtime |

## What does text to speech do?

Text to speech takes a script and reads it aloud in a voice you choose. Paxa TTS Flash v2 reads English, Thai, and Mandarin, with 50 voices to choose from. People use it for narration, voice-overs, audiobooks, announcements, and the voice of an assistant. [What is text to speech](https://paxalabs.com/guides/what-is-text-to-speech) explains how it works.

## What does speech to text do?

Speech to text takes audio and writes down the words. Paxa STT Lite transcribes a recorded file, with speaker numbers and caption files when you ask for them. Paxa STT Lite Realtime is streaming transcription with automatic turn detection, for live audio. Both are multilingual. [How to transcribe audio to text](https://paxalabs.com/guides/how-to-transcribe-audio-to-text) walks through a first file.

## Which one do you need?

| You want to | Use |
| --- | --- |
| Turn a script into a voice-over or an audiobook | Text to speech |
| Read articles or lessons aloud | Text to speech |
| Give an app or a phone line a voice | Text to speech |
| Get minutes from a meeting recording | Speech to text |
| Caption a video | Speech to text |
| Let people dictate or talk to an app | Speech to text |

## When do you need both?

Some products chain the two. A voice agent transcribes what the caller says, decides on a reply, and speaks it. Dubbing transcribes a video, translates the transcript, and voices it in another language. [What an AI voice agent is](https://paxalabs.com/guides/what-is-an-ai-voice-agent) and [the dubbing page](https://paxalabs.com/dubbing) show each one.

## What does each one cost?

Text to speech is priced by the characters of text, and speech to text by the minutes of audio. Realtime transcription has its own rate, billed on the audio received.

| Model | Credits | Money |
| --- | --- | --- |
| Paxa TTS Flash v2 | 10 per 1,000 characters | $10 per million characters |
| Paxa STT Lite | 8.33 per minute | $0.50 per hour |
| Paxa STT Lite Realtime | 12.5 per minute | $0.75 per hour |

A new account gets 100 free credits with no card, to spend on either. They are for personal use and evaluation. Buy any plan or credit pack, and the account's audio and transcripts are yours to use commercially, including those made before the purchase. [The Terms](https://paxalabs.com/terms#content-and-output) set this out.

## What other names do they go by?

Text to speech is also called TTS, speech synthesis, and AI voice generation. Speech to text is also called STT, transcription, automatic speech recognition, and ASR. Each name has its own page: [speech synthesis](https://paxalabs.com/speech-synthesis), [the AI voice generator](https://paxalabs.com/ai-voice-generator), [ASR](https://paxalabs.com/asr), and [online transcription](https://paxalabs.com/transcribe).

## What should you read or try next?

- Hear a script read aloud in [the Speech tool](https://paxalabs.com/app/speech).
- Upload a recording to [the Transcribe tool](https://paxalabs.com/app/transcribe).
- Try [free text to speech online](https://paxalabs.com/tts) in English, Thai, and Mandarin.
- Read about each product on [the text to speech page](https://paxalabs.com/text-to-speech) and [the speech to text page](https://paxalabs.com/speech-to-text).
- Compare providers on [the text to speech API comparison](https://paxalabs.com/compare/text-to-speech-apis) and [the speech to text API comparison](https://paxalabs.com/compare/speech-to-text-apis).

## Common questions

### Is speech recognition the same as voice recognition?

They are different jobs. Speech recognition, another name for speech to text, writes down what was said. Voice recognition identifies who is speaking. Paxa's speaker numbers tell the voices in one recording apart and identify nobody.

### Can one platform do both?

Yes. Paxa offers text to speech and speech to text on one account and one credit balance, and the console has a tool for each.

### Can I try them for free?

Yes. A new account's 100 free credits work in both tools, with no card. On transcription alone they cover about 12 minutes of audio.

### Which languages do they cover?

Paxa TTS Flash v2 reads English, Thai, and Mandarin. Transcription is multilingual and detects the language for you.