What is Lexora Speech?
Lexora Speech is the generation layer responsible for converting text into natural-sounding audio using advanced neural voice models.
Users simply write their text, choose a language and a voice, then generate the audio in seconds.
- AI Text-to-Speech (TTS): converts written content into realistic spoken audio.
- Multilingual support: generate speech in multiple languages by selecting the desired language.
- Neural voices: choose from a wide range of voices with different tones and styles.
- Credit-based generation: usage is calculated based on the amount of generated audio.
Language and Voice Selection
Before generating audio, you choose the language and voice that will be used during speech synthesis.
These selections determine how the text will be interpreted by the speech engine, including pronunciation rules, phonetic modeling, and vocal tone.
How it works
- Select the language that matches your text.
- Choose the voice style you want to use.
- Click Generate to start the rendering process.
This approach gives you full control over the voice style and language used for each audio generation.
Speech Generation Workflow
Once you press Generate, Lexora runs a streamlined pipeline to convert your text into speech.
- Text validation: ensures the content is valid and processable.
- Credit estimation: calculates the required credits.
- Voice rendering: generates the speech waveform using neural models.
- Audio asset creation: stores the generated audio.
- Audio ID creation: assigns a unique identifier for reuse and embedding.
Once created, the audio asset can be reused multiple times without regenerating it.
Voice Selection
Different voices offer different tones, pacing, and personality. Choosing the right voice helps match the delivery to your content.
- Narrative voices: ideal for storytelling and long-form articles.
- Neutral voices: suitable for documentation and informational content.
- Dynamic voices: useful for announcements and promotional material.
The selected voice and language work together to produce accurate pronunciation and natural intonation.
Credits and Usage Model
Lexora Speech operates on a transparent credit-based system. Before generating audio, the platform estimates the required credits so you always know the cost in advance.
- No generation starts without sufficient credits.
- Credits are deducted only when audio is successfully rendered.
- The system scales from small blogs to large platforms.
For detailed information, see the Credits documentation.
Common Use Cases
- Transform blog posts into audio versions.
- Improve accessibility for users who prefer listening.
- Create multilingual audio content.
- Build scalable audio layers for content platforms.