> ## Documentation Index > Fetch the complete documentation index at: https://docs.modellix.ai/llms.txt > Use this file to discover all available pages before exploring further. # MiniMax Speech 2.8 HD > [Core Function] MiniMax Speech 2.8 HD is a high-quality text-to-speech model that converts text into natural spoken audio, including expressive paralinguistic cues such as (laughs) and (sighs). [Strengths] Strong narration quality, stable prosody controls (speed, volume, pitch, emotion), optional timbre mixing, pronunciation overrides, subtitles, and flexible audio formats. [Best For] Highly recommended for: brand voiceovers, audiobook or long-form narration, marketing clips, multilingual delivery with language_boost, and expressive character speech. [Limitations] Do NOT use this for streaming or real-time partial audio. Do NOT request hex output or emotion whisper. Do NOT send both voice_id and timbre_weights. Audio URLs expire in about 24 hours. [Routing] Choose speech-2.8-hd when quality or expressive delivery matters most. Choose speech-2.8-turbo when the user prioritizes lower latency or cost. ## OpenAPI ````yaml /model-api/minimax/minimax-t2s.json post /minimax/speech-2.8-hd openapi: 3.1.0 info: title: MiniMax Text-to-Speech API description: >- MiniMax Speech 2.8 text-to-speech (speech-2.8-hd and speech-2.8-turbo). Submit an async task, then poll GET /api/v1/tasks/{task_id} for the audio URL (approximately 24-hour validity). Streaming and hex audio output are not supported. version: 1.0.0 contact: name: Modellix Support email: support@modellix.ai servers: - url: https://api.modellix.ai/api/v1 description: Modellix API security: - bearerAuth: [] paths: /minimax/speech-2.8-hd: post: summary: MiniMax Speech 2.8 HD description: >- [Core Function] MiniMax Speech 2.8 HD is a high-quality text-to-speech model that converts text into natural spoken audio, including expressive paralinguistic cues such as (laughs) and (sighs). [Strengths] Strong narration quality, stable prosody controls (speed, volume, pitch, emotion), optional timbre mixing, pronunciation overrides, subtitles, and flexible audio formats. [Best For] Highly recommended for: brand voiceovers, audiobook or long-form narration, marketing clips, multilingual delivery with language_boost, and expressive character speech. [Limitations] Do NOT use this for streaming or real-time partial audio. Do NOT request hex output or emotion whisper. Do NOT send both voice_id and timbre_weights. Audio URLs expire in about 24 hours. [Routing] Choose speech-2.8-hd when quality or expressive delivery matters most. Choose speech-2.8-turbo when the user prioritizes lower latency or cost. operationId: minimaxSpeech28Hd requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/MiniMaxSpeechT2SRequest' examples: basic: summary: Minimal single-voice synthesis value: text: Hello, this is MiniMax speech synthesis. voice_id: English_expressive_narrator with_controls: summary: Prosody, format, and language boost value: text: There is a large garden behind my house. voice_id: English_expressive_narrator format: mp3 sample_rate: 32000 speed: 1 vol: 1 pitch: 0 emotion: calm language_boost: English timbre_mix: summary: Mixed timbres (no top-level voice_id) value: text: Mixed timbre demo. timbre_weights: - voice_id: Chinese (Mandarin)_Lyrical_Voice weight: 30 - voice_id: Chinese (Mandarin)_News_Anchor weight: 70 responses: '200': description: >- Task submitted successfully. Poll GET /api/v1/tasks/{task_id} until the task completes; synthesized audio is returned on the task result. content: application/json: schema: $ref: '#/components/schemas/AsyncTaskResponse' example: code: 0 message: success data: status: pending task_id: task-speech-28-hd-001 model_id: minimax/speech-2.8-hd get_result: method: GET url: https://api.modellix.ai/api/v1/tasks/task-speech-28-hd-001 '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '429': $ref: '#/components/responses/TooManyRequests' '500': $ref: '#/components/responses/InternalServerError' components: schemas: MiniMaxSpeechT2SRequest: type: object required: - text description: >- text is always required. Exactly one of voice_id or timbre_weights is required (mutually exclusive). stream, stream_options, and output_format are not accepted. oneOf: - required: - voice_id title: single_voice description: >- Single-voice synthesis: provide voice_id; do not provide timbre_weights. - required: - timbre_weights title: timbre_mix description: >- Mixed-voice synthesis: provide timbre_weights; do not provide voice_id. properties: text: type: string minLength: 1 maxLength: 9999 description: >- Text to synthesize (fewer than 10000 characters). Supports pause markers such as <#0.5#>, inline pronunciation in parentheses, and paralinguistic tags such as (laughs) or (sighs). example: Hello, this is MiniMax speech synthesis. voice_id: type: string minLength: 1 maxLength: 256 description: >- Required when not using timbre_weights (exactly one of voice_id or timbre_weights). Accepts a system, cloned, or AI-generated voice ID. Mutually exclusive with timbre_weights. Full official system voice catalog: https://platform.minimax.io/docs/faq/system-voice-id . Sample system voices (not exhaustive): English_expressive_narrator, English_radiant_girl, English_magnetic_voiced_man, Chinese (Mandarin)_News_Anchor, Chinese (Mandarin)_Lyrical_Voice, Chinese (Mandarin)_Reliable_Executive. example: English_expressive_narrator x-systemVoiceIds: - English_expressive_narrator - English_radiant_girl - English_magnetic_voiced_man - Chinese (Mandarin)_News_Anchor - Chinese (Mandarin)_Lyrical_Voice - Chinese (Mandarin)_Reliable_Executive timbre_weights: type: array minItems: 1 maxItems: 4 description: >- Required when not using voice_id (exactly one of voice_id or timbre_weights). Mix up to 4 voices by weight. Mutually exclusive with top-level voice_id. items: $ref: '#/components/schemas/MiniMaxTimbreWeight' speed: type: number minimum: 0.5 maximum: 2 default: 1 description: Speech rate multiplier. Higher is faster. example: 1 vol: type: number exclusiveMinimum: 0 maximum: 10 default: 1 description: Speech volume. Range (0, 10]. example: 1 pitch: type: integer minimum: -12 maximum: 12 default: 0 description: >- Pitch adjustment in semitones. Distinct from modify_pitch (voice effect slider). example: 0 emotion: type: string enum: - happy - sad - angry - fearful - disgusted - surprised - calm - fluent description: >- Optional emotion control. Leave unset to let the model choose naturally. whisper is not supported on speech-2.8-hd or speech-2.8-turbo. example: calm text_normalization: type: boolean default: false description: >- Enable Chinese/English text normalization for better number reading; may add slight latency. example: false latex_read: type: boolean default: false description: >- Enable LaTeX formula reading (Chinese). When enabled, language_boost may be forced to Chinese. Wrap formulas with $$. example: false sample_rate: type: integer enum: - 8000 - 16000 - 22050 - 24000 - 32000 - 44100 default: 32000 description: Output sample rate in Hz. example: 32000 bitrate: type: integer enum: - 32000 - 64000 - 128000 - 256000 default: 128000 description: Audio bitrate. Applies to mp3 only. example: 128000 format: type: string enum: - mp3 - pcm - flac - wav - pcmu_raw - pcmu_wav - opus default: mp3 description: Output audio format. pcmu_raw and pcmu_wav are G.711 mu-law (8 kHz). example: mp3 channel: type: integer enum: - 1 - 2 default: 1 description: Number of channels. 1 is mono, 2 is stereo. example: 1 force_cbr: type: boolean default: false description: >- Force constant bitrate encoding. Optional; typically unused for standard async synthesis. example: false pronunciation_tone: type: array items: type: string description: >- Custom pronunciation rules in original/replacement form, for example omg/oh my god or Chinese pinyin with tone numbers. example: - omg/oh my god language_boost: type: string enum: - 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 description: >- Boost recognition for a language or dialect. Use auto when the language is unknown. example: English modify_pitch: type: integer minimum: -100 maximum: 100 description: >- Voice effect: deepen (negative) to brighten (positive). Distinct from pitch (semitone). example: 0 modify_intensity: type: integer minimum: -100 maximum: 100 description: 'Voice effect: stronger (negative) to softer (positive).' example: 0 modify_timbre: type: integer minimum: -100 maximum: 100 description: 'Voice effect: fuller (negative) to crisper (positive).' example: 0 sound_effects: type: string enum: - spacious_echo - auditorium_echo - lofi_telephone - robotic description: Optional single sound effect applied to the output. example: spacious_echo subtitle_enable: type: boolean default: false description: Whether to generate a subtitle file download link. example: false subtitle_type: type: string enum: - sentence - word default: sentence description: Subtitle granularity. word_streaming is not supported. example: sentence AsyncTaskResponse: description: Response object for asynchronous task submission. type: object required: - code - message - data properties: code: type: integer description: Response code, 0 indicates success example: 0 message: type: string description: Response message example: success data: type: object required: - status - task_id - model_id description: >- Task submission details. Poll GET /api/v1/tasks/{task_id} until status is success; audio appears in result.resources. properties: status: type: string enum: - pending - processing description: Initial task status example: pending task_id: type: string description: Unique task identifier for polling example: task-speech-28-hd-001 model_id: type: string description: Model ID in provider/model format example: minimax/speech-2.8-hd get_result: type: object description: >- Endpoint to query the task result. See Common API: Query Task Result. properties: method: type: string example: GET url: type: string example: https://api.modellix.ai/api/v1/tasks/task-speech-28-hd-001 MiniMaxTimbreWeight: type: object required: - voice_id - weight properties: voice_id: type: string minLength: 1 maxLength: 256 description: >- Voice ID for one entry in a timbre mix. Distinct from the top-level single-voice voice_id field. May be a system voice from the official catalog (https://platform.minimax.io/docs/faq/system-voice-id) or a cloned/AI voice ID. example: female-chengshu weight: type: integer minimum: 1 maximum: 100 description: >- Relative mix weight for this voice. Higher values increase similarity to this voice. At most 4 voices. example: 30 ErrorResponse: type: object required: - code - message properties: code: type: integer example: 400 message: type: string example: 'Invalid parameters: voice_id or timbre_weights is required' responses: BadRequest: description: Invalid request parameters content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' Unauthorized: description: Missing or invalid API key content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' TooManyRequests: description: Rate limit exceeded content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' InternalServerError: description: Internal server error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API Key ````