infery
← All models

Tada

tada-3b-text-to-speech

Text to Speechby Tada

A unified speech-language model that synchronizes speech and text into a single, cohesive stream via 1:1 alignment.

Details

Accepts
audio

Pricing

Price
0.010 cr / character

Prices in credits (1 credit = $0.01).

Data schema

Input

FieldTypeDescription
top_pnumberTop-p (nucleus) sampling parameter for text generation.
promptrequiredstringThe text to synthesize into speech using the reference speaker's voice.
languagestringenum: en, ar, ch, de, es, frLanguage for text alignment. Use the appropriate code for non-English synthesis.
audio_urlrequiredstring
transcriptstringTranscript of the reference audio. For non-English audio, providing a transcript is required since the built-in ASR is English-only.
temperaturenumberSampling temperature for text token generation. Must be greater than 0. Higher values produce more varied output.
output_formatstringenum: wav, mp3The format of the output audio file.
num_extra_stepsintegerNumber of extra autoregressive steps for speech continuation beyond the input text. Useful for generating trailing prosody or silence.
speed_up_factornumberFactor to speed up or slow down the generated speech. Values > 1.0 speed up, < 1.0 slow down.
noise_temperaturenumberTemperature for noise in the flow matching diffusion process.
acoustic_cfg_scalenumberClassifier-free guidance scale for acoustic feature generation.
repetition_penaltynumberPenalty applied to repeated tokens during generation.
num_inference_stepsintegerNumber of ODE solver steps for flow matching acoustic generation. More steps improve quality at the cost of speed.

Output

FieldTypeDescription
audioThe generated speech audio file.