joshu

MiniMax: Speech 2.8 Turbo

From MiniMax, through Joshu Data Bodega.

Part of Voice generation speech 2 8 turbo generate.

Price
$0.000066 per call$0.07 per 1,000 calls

No calls measured in the last 30 days.

Prices include the router's 10% markup and are paid from the usage wallet. Checked on November 11, 2026.

Inputs

  • textstring, required

    Fewer than 10,000 characters. Use #x# between speakable segments for a pause of x seconds.

  • modelstring: speech-2.8-turbo, required

  • streamboolean: false, required

    This catalog route is non-streaming so the relay returns one bounded JSON response.

  • voice_modifyobject, optional

    Optional pitch, intensity, timbre and sound_effects controls.

  • audio_settingobject, optional

    Non-streaming output supports mp3, wav and flac. bitrate (mp3 only) is 32000 | 64000 | 128000 | 256000; 192000 is invalid (status 2013). sample_rate is 8000 | 16000 | 22050 | 24000 | 32000 | 44100.

  • output_formatstring: url, required

    The returned audio URL is valid for 24 hours.

  • subtitle_typestring: sentence, word, optional

  • voice_settingobject, required

    voice_id is required. Optional controls include speed, vol, pitch, emotion, text_normalization and latex_read. speech-2.8 emotion is happy | sad | angry | fearful | disgusted | surprised | calm; whisper and fluent are 2.6-only (status 2013 on 2.8, even with a whispering-named voice_id — use a whispering-style voice_id instead, or speech-2.6 for emotion=whisper).

  • language_booststring: Chinese, Chinese,Yue, English, Arabic, Russian, Spanish, French, Portuguese, German, Turkish, Dutch, Ukrainian, Vietnamese, Indonesian, Japanese, Italian, Korean, Thai, Polish, Romanian, Greek, Czech, Finnish, Hindi, Bulgarian, Danish, Hebrew, Malay, Persian, Slovak, Swedish, Croatian, Filipino, Hungarian, Norwegian, Slovenian, Catalan, Nynorsk, Tamil, Afrikaans, auto, optional

    Exact enum strings only. English(UK) and en-GB are invalid (status 2013); use English. auto detects the language.

  • subtitle_enableboolean, optional

  • pronunciation_dictobject, optional

    Custom pronunciation pairs in source/target form.

Call it

Router call with a router API key
curl https://hello.joshu.me/api/router/call \
  -H "Authorization: Bearer $JOSHU_ROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "endpoint": "minimax.voice-gen.speech-2-8-turbo",
    "params": {
      "text": "Hello.",
      "model": "speech-2.8-turbo",
      "stream": false,
      "audio_setting": {
        "format": "mp3",
        "bitrate": 128000,
        "channel": 1,
        "sample_rate": 32000
      },
      "output_format": "url",
      "voice_setting": {
        "vol": 1,
        "pitch": 0,
        "speed": 1,
        "voice_id": "<voice ID>"
      }
    }
  }'