MiniMax: Speech 2.8 HD

Name: Speech 2.8 HD
Author: MiniMax

minimax/speech-2-8-hd

active tts 2 providers

Playground

MiniMax Speech 2.8 HD text-to-speech model for higher-quality speech synthesis with streaming, subtitle, language boost, voice modification, and pronunciation controls.

AuthorMiniMax

Languages24 languages

VoicesLoading...

Response formatsMP3 · PCM · FLAC · WAV · µ-law · Opus

Statusactive

Quick start

Speech 2.8 HD

Create an API key

Create API key

Make a request

Pick an SDK, request mode, and language. The model and provider route are already filled in.

ElevenLabs SDK streaming TTS

Routes the ElevenLabs SDK through allmodels and streams the audio response.

import { ElevenLabsClient } from "elevenlabs";

const client = new ElevenLabsClient({
  apiKey: process.env.ALLMODELS_API_KEY,
  baseUrl: "https://api.allmodels.io/el"
});

const stream = await client.textToSpeech.stream("voice_id", {
  text: "I'm sorry, Dave. I'm afraid I can't do that.",
  modelId: "minimax/speech-2-8-hd",
  outputFormat: "mp3_44100_128"
}, {
  queryParams: {
    provider: {
      "only": [
        "minimax"
      ]
    }
  }
});

const reader = stream.getReader();
for (;;) {
  const { done, value } = await reader.read();
  if (done) break;
  // Write value to your response, file, or audio playback pipeline.
}

Pass provider parameters

Pass provider preferences through the compatible SDK request options when you want to pin a route to a provider.

{
  queryParams: {
    provider: {
      only: ["minimax"]
    }
  }
}

Providers

Different providers can host the same model. Choose one provider when you need a fixed backend, or let routing select among them.

ProviderSlugPer 1k charactersLatencyiModesUptime

MiniMax 13 params

minimax

$0.1

—

Synchronous · Streaming

MiniMax

13 documented provider parameters for this route.

Pricing

ModeLanguagePer 1k characters

HD Any $0.1

Parameters

model body · string · required

The speech synthesis model version to use.

8 enum values

Values

speech-2.8-hdspeech-2.8-turbospeech-2.6-hdspeech-2.6-turbospeech-02-hdspeech-02-turbospeech-01-hdspeech-01-turbo

text body · string · required

The text to be converted into speech. Must be less than 10,000 characters. For texts over 3,000 characters, streaming output is recommended. Paragraph breaks should be marked with newline characters. Pause control markers may be added in the form <#x#>, where x is the pause duration in seconds with valid range [0.01, 99.99], up to two decimal places. Pause markers must be placed between speakable text segments and cannot be used consecutively. Inline pronunciation supports Mandarin Pinyin with tone number 1–5, IPA symbols, or Cantonese Jyutping with tone number 1–6 in half-width parentheses. Interjection tags are only supported when using speech-2.8-hd or speech-2.8-turbo models. Supported interjection tags include: (laughs), (chuckle), (coughs), (clear-throat), (groans), (breath), (pant), (inhale), (exhale), (gasps), (sniffs), (sighs), (snorts), (burps), (lip-smacking), (humming), (hissing), (emm), and (sneezes).

max 9999

audio_setting body · object · optional

Audio setting object. Child attributes are referenced in the documentation but not fully included in the provided source text.

language_boost body · string | null · optional

Controls whether recognition for specific minority languages and dialects is enhanced. If the language type is unknown, set to "auto" and the model will automatically detect it. The speech-01 and speech-02 series models do not currently support Persian, Filipino, or Tamil.

default null42 enum values

Values

ChineseChinese,YueEnglishArabicRussianSpanishFrenchPortugueseGermanTurkishDutchUkrainianVietnameseIndonesianJapaneseItalianKoreanThaiPolishRomanianGreekCzechFinnishHindiBulgarianDanishHebrewMalayPersianSlovakSwedishCroatianFilipinoHungarianNorwegianSlovenianCatalanNynorskTamilAfrikaansauto

output_format body · string · optional

Controls the output format. Only effective in non-streaming scenarios. In streaming, only hex is supported. Returned URLs are valid for 24 hours.

default "hex"2 enum values

Values

urlhex

pronunciation_dict body · object · optional

Pronunciation dictionary object. Child attributes are referenced in the documentation but not fully included in the provided source text.

stream body · boolean · optional

Whether to enable streaming output.

default false

stream_options body · object · optional

Stream options object. Child attributes are referenced in the documentation but not included in the provided source text.

subtitle_enable body · boolean · optional

Controls whether subtitles are enabled. Available for models: speech-2.8-hd, speech-2.8-turbo, speech-2.6-hd, speech-2.6-turbo, speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo.

default false

subtitle_type body · string · optional

Subtitle granularity. sentence: sentence-level timestamps. word: word-level timestamps. word_streaming: word-level timestamps optimized for streaming, only valid when stream=true.

default "sentence"3 enum values

Values

sentencewordword_streaming

timbre_weights body · array · optional

Timbre weights legacy field. Child attributes are referenced in the documentation but not included in the provided source text.

voice_modify body · object · optional

Voice effects configuration. Supported audio formats: non-streaming mp3, wav, flac; streaming mp3.

voice_setting body · object · optional

Voice setting object. Child attributes are referenced in the documentation but not fully included in the provided source text.

Fal 8 params

fal

$0.1

—

Synchronous

Fal

8 documented provider parameters for this route.

Pricing

ModeLanguagePer 1k characters

$ / 1k Characters Any $0.1

Parameters

prompt body · string · required

Text to convert to speech. Supports pause markers like <#0.1#> and interjection tags such as (laughs), (sighs), (coughs), (clears throat), (gasps), (sniffs), (groans), and (yawns).

audio_setting body · object · optional

Audio configuration settings for FAL MiniMax Speech 2.8 endpoints.

language_boost body · string · optional

Enhance recognition of specified languages and dialects.

38 enum values

Values

ChineseChinese,YueEnglishArabicRussianSpanishFrenchPortugueseGermanTurkishDutchUkrainianVietnameseIndonesianJapaneseItalianKoreanThaiPolishRomanianGreekCzechFinnishHindiBulgarianDanishHebrewMalaySlovakSwedishCroatianHungarianNorwegianSlovenianCatalanNynorskAfrikaansauto

normalization_setting body · object · optional

Loudness normalization settings for the audio.

output_format body · string · optional

Format of the output content for non-streaming requests.

default "hex"2 enum values

Values

urlhex

pronunciation_dict body · object · optional

Custom pronunciation dictionary for text replacement.

voice_modify body · object · optional

Voice modification settings to adjust pitch, intensity, and timbre.

voice_setting body · object · optional

Voice configuration settings for FAL MiniMax Speech 2.8 endpoints.

Supported languages

24 languages

Arabic ar

Cantonese yue

Chinese (Mandarin) zh

Czech cs

Dutch nl

English en

Finnish fi

French fr

German de

Greek el

Hindi hi

Indonesian id

Italian it

Japanese ja

Korean ko

Polish pl

Portuguese pt

Romanian ro

Russian ru

Spanish es

Thai th

Turkish tr

Ukrainian uk

Vietnamese vi

Voices

Features

Text to speech

Generates spoken audio from text input.

text_to_speech

Streaming

Supports streaming output with stream=true.

streaming

Synchronous

Synthesizes speech from fully-submitted text in a single synchronous request.

synchronous

Subtitles

Supports sentence, word, and streaming word subtitle modes.

subtitles

Language boost

Supports language-specific recognition enhancement or automatic language detection with language_boost=auto.

language_boost

Voice modification

Supports pitch, intensity, timbre, and sound effect controls through voice_modify.

voice_modification

Pronunciation dictionary

Supports pronunciation replacement through pronunciation_dict.

pronunciation_dictionary

Code

ElevenLabs SDK streaming TTS

Routes the ElevenLabs SDK through allmodels and streams the audio response.

import { ElevenLabsClient } from "elevenlabs";

const client = new ElevenLabsClient({
  apiKey: process.env.ALLMODELS_API_KEY,
  baseUrl: "https://api.allmodels.io/el"
});

const stream = await client.textToSpeech.stream("voice_id", {
  text: "I'm sorry, Dave. I'm afraid I can't do that.",
  modelId: "minimax/speech-2-8-hd",
  outputFormat: "mp3_44100_128"
}, {
  queryParams: {
    provider: {
      "only": [
        "minimax"
      ]
    }
  }
});

const reader = stream.getReader();
for (;;) {
  const { done, value } = await reader.read();
  if (done) break;
  // Write value to your response, file, or audio playback pipeline.
}