alibaba/cosyvoice-v3-flash

cosyvoice-v3-flash
文档
Schema

[Core Function] CosyVoice v3 Flash is Alibaba's low-latency text-to-speech model. [Strengths] Rich system voice catalog, fixed-format instruction on Instruct-capable system voices, SSML, hot_fix, AIGC watermark, Markdown filter (cloned voices only), and multiple audio formats with faster turnaround than Plus. [Best For] Voice assistants, interactive prompts, IVR, short announcements, dialect system voices, and latency-sensitive batch TTS. [Limitations] Do NOT use when maximum speech quality or long-form audiobook fidelity is the priority (use Plus). System-voice instruction must follow CosyVoice voice-list fixed Chinese formats. [Routing] Choose Flash when speed or a richer system-voice catalog matters most. Choose Plus for premium narration quality.

$8.0637/M chars
text-to-speech

输入

Text to synthesize. Required. Max 20,000 Unicode characters. Supports plain text, SSML (set enable_ssml to true), and LaTeX formulas per Alibaba documentation.
System or cloned voice ID for cosyvoice-v3-flash. Full official CosyVoice voice catalog (use only voices listed under cosyvoice-v3-flash for system voices): https://help.aliyun.com/zh/model-studio/cosyvoice-voice-list . Do not mix Plus and Flash system voices. Cloned or designed voice IDs are also accepted. Sample system voices (not exhaustive): longanyang, longhuhu_v3, longpaopao_v3, longshanshan_v3, longjiaxin_v3, loongkyong_v3, longfei_v3, longyingxiao_v3, longyingxun_v3, longxiaochun_v3, longantai_v3, longmiao_v3, longjiqi_v3, longanran_v3, longshuo_v3.
AIGC PropagateID. Only effective when enable_aigc_tag is true.
AIGC ContentPropagator. Only effective when enable_aigc_tag is true.
Text hot-fix before synthesis. Optional object with pronunciation (custom pinyin for Chinese words) and replace (text substitution) arrays per Alibaba HTTP API.
Optional speaking-style instruction for Instruct-capable system voices. Enforced weighted length ≤100 (CJK ideographs / Han count as 2; other characters count as 1). Alibaba requires fixed Chinese formats from the CosyVoice voice list for system voices.
Audio bit rate in kbps. Optional. Range 6 to 510. Only supported when format is opus; do not use for mp3, pcm, or wav.
Embed AIGC invisible watermark into wav/mp3/opus output. Supported on cosyvoice-v3-plus and cosyvoice-v3-flash.
Whether to parse text as SSML. Default false. When true, text must follow Alibaba CosyVoice SSML rules.
Audio encoding format. Default mp3.
mp3
Target language hint for pronunciation (numbers, symbols, minor languages).
language_hint
Pitch multiplier. Default 1.0. Range 0.5 (lower) to 2.0 (higher).
Speech rate multiplier. Default 1.0. Range 0.5 (slow) to 2.0 (fast).
Audio sample rate in Hz. Default 22050.
22050
Random seed for reproducible synthesis when text, voice, and other parameters are identical. Default 0. Range 0 to 65535.
Output volume. Default 50. Range 0 (silent) to 100 (maximum).
Filter Markdown markers before synthesis. Only supported on cosyvoice-v3-flash with cloned voices.

结果

暂无结果

运行模型后，结果将在这里显示。

Next:

README

属性

系列

参数

参数名 描述 类型 必填 枚举值
text Text to synthesize. Required. Max 20,000 Unicode characters. Supports plain text, SSML (set enable_ssml to true), and LaTeX formulas per Alibaba documentation. string 是 -
voice System or cloned voice ID for cosyvoice-v3-flash. Full official CosyVoice voice catalog (use only voices listed under cosyvoice-v3-flash for system voices): https://help.aliyun.com/zh/model-studio/cosyvoice-voice-list . Do not mix Plus and Flash system voices. Cloned or designed voice IDs are also accepted. Sample system voices (not exhaustive): longanyang, longhuhu_v3, longpaopao_v3, longshanshan_v3, longjiaxin_v3, loongkyong_v3, longfei_v3, longyingxiao_v3, longyingxun_v3, longxiaochun_v3, longantai_v3, longmiao_v3, longjiqi_v3, longanran_v3, longshuo_v3. string 是 -
aigc_propagate_id AIGC PropagateID. Only effective when enable_aigc_tag is true. string 否 -
aigc_propagator AIGC ContentPropagator. Only effective when enable_aigc_tag is true. string 否 -
hot_fix Text hot-fix before synthesis. Optional object with pronunciation (custom pinyin for Chinese words) and replace (text substitution) arrays per Alibaba HTTP API. object 否 -
instruction Optional speaking-style instruction for Instruct-capable system voices. Enforced weighted length ≤100 (CJK ideographs / Han count as 2; other characters count as 1). Alibaba requires fixed Chinese formats from the CosyVoice voice list for system voices. string 否 -
bit_rate Audio bit rate in kbps. Optional. Range 6 to 510. Only supported when format is opus; do not use for mp3, pcm, or wav. integer 否 -
enable_aigc_tag Embed AIGC invisible watermark into wav/mp3/opus output. Supported on cosyvoice-v3-plus and cosyvoice-v3-flash. boolean 否 true, false
enable_ssml Whether to parse text as SSML. Default false. When true, text must follow Alibaba CosyVoice SSML rules. boolean 否 true, false
format Audio encoding format. Default mp3. string 否 mp3, pcm, wav, opus
language_hint Target language hint for pronunciation (numbers, symbols, minor languages). string 否 zh, en, fr, de, ja, ko, ru, pt, th, id, vi, es, it, ms, fil, ar
pitch Pitch multiplier. Default 1.0. Range 0.5 (lower) to 2.0 (higher). number 否 -
rate Speech rate multiplier. Default 1.0. Range 0.5 (slow) to 2.0 (fast). number 否 -
sample_rate Audio sample rate in Hz. Default 22050. integer 否 8000, 16000, 22050, 24000, 44100, 48000
seed Random seed for reproducible synthesis when text, voice, and other parameters are identical. Default 0. Range 0 to 65535. integer 否 -
volume Output volume. Default 50. Range 0 (silent) to 100 (maximum). integer 否 -
enable_markdown_filter Filter Markdown markers before synthesis. Only supported on cosyvoice-v3-flash with cloned voices. boolean 否 true, false

价格

单位: $/M chars

价格
$8.0637/M chars

相关模型

  • alibaba/cosyvoice-design: [Core Function] CosyVoice Design creates a temporary voice from a natural-language voice_prompt and synthesizes speech in one async request; only the final audio is returned. [Strengths] No voice enrollment management; language_hint (zh/en) applies to both design and synthesis; same prosody and format controls as CosyVoice TTS. [Best For] One-off designed speech without storing enrolled voices. [Limitations] Do NOT use this to obtain a reusable voice library entry; the designed voice is temporary and is not returned. model must be cosyvoice-v3.5-plus or cosyvoice-v3.5-flash; voice_prompt and text are required.
  • alibaba/cosyvoice-v3-plus: [Core Function] CosyVoice v3 Plus is Alibaba’s high-quality text-to-speech model. [Strengths] System voices (e.g. longanyang, longanhuan), SSML and LaTeX input, hot_fix pronunciation correction, AIGC watermark, and output in mp3, pcm, wav, or opus. System-voice instruction must use the fixed Chinese formats in the CosyVoice voice list. [Best For] Brand voiceovers, audiobooks, high-quality narration, marketing clips, and scenarios where speech quality matters more than minimum latency. [Limitations] Do NOT use for real-time streaming or word-level timestamps. Fewer system voices than Flash. [Routing] Choose Plus when quality or narration fidelity matters most. Choose cosyvoice-v3-flash for lower latency or a richer system-voice catalog.
  • alibaba/cosyvoice-clone: [Core Function] CosyVoice Clone clones a speaker from a public reference audio URL and synthesizes new speech in one async request; only the final audio is returned. [Strengths] No voice enrollment management; language_hint applies to both cloning and synthesis; SSML, hot_fix, and prosody controls. [Best For] One-off cloned narration and demos where a lasting voice library is not needed. [Limitations] Do NOT use this to obtain a reusable voice library entry; the cloned voice is temporary and is not returned. Reference URL must be publicly accessible. model must be cosyvoice-v3.5-plus or cosyvoice-v3.5-flash. [Routing] Choose cosyvoice-v3.5-plus for higher speech quality; cosyvoice-v3.5-flash for lower latency.

相关资源

CosyVoice
CosyVoiceSeries
alibaba/cosyvoice-design
alibaba/cosyvoice-designtext-to-speech
alibaba/cosyvoice-v3-plus
alibaba/cosyvoice-v3-plustext-to-speech
alibaba/cosyvoice-clone
alibaba/cosyvoice-clonespeech-to-speech