> ## Documentation Index > Fetch the complete documentation index at: https://docs.aihubmax.com/llms.txt > Use this file to discover all available pages before exploring further. # Paraformer V2 语音识别 > - Paraformer V2 录音文件识别模型，支持中文（含方言）、英文、日语等多语种 - 支持说话人分离、热词定制、语气词过滤 - 异步处理模式，使用返回的任务ID [进行查询](/pages/zh/api-manual/task-management/get-task-detail) - 识别结果会在任务详情的 `results` 字段中返回 ## OpenAPI ````yaml openapi/zh/paraformer-v2.json POST /v1/audios/generations openapi: 3.1.0 info: title: Paraformer V2 语音识别 version: '1.0' servers: - url: https://api.aihubmax.com security: - BearerAuth: [] paths: /v1/audios/generations: post: tags: - Audio > 语音识别 summary: Paraformer V2 语音识别 description: > - Paraformer V2 录音文件识别模型，支持中文（含方言）、英文、日语等多语种 - 支持说话人分离、热词定制、语气词过滤 - 异步处理模式，使用返回的任务ID [进行查询](/pages/zh/api-manual/task-management/get-task-detail) - 识别结果会在任务详情的 `results` 字段中返回 operationId: paraformer-v2 requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ParaformerRequest' examples: basic_transcription_with_default_settings: summary: Basic transcription with default settings value: model: paraformer-v2 file_urls: - https://example.com/audio/meeting.wav multi_language_transcription_with_diarization: summary: Multi-language transcription with diarization value: model: paraformer-v2 file_urls: - https://example.com/audio/interview_part1.mp3 - https://example.com/audio/interview_part2.mp3 language_hints: - zh - en diarization: enabled: true speaker_count: 3 8khz_telephone_audio_with_recognition_tuning: summary: 8kHz telephone audio with recognition tuning value: model: paraformer-8k-v2 file_urls: - https://example.com/audio/call_recording.wav channel_id: - 0 - 1 recognition: disfluency_removal_enabled: true vocabulary_id: vocab-xxxx responses: '200': $ref: '#/components/responses/TaskCreated' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '402': $ref: '#/components/responses/PaymentRequired' '422': $ref: '#/components/responses/UnprocessableEntity' '429': $ref: '#/components/responses/TooManyRequests' '500': $ref: '#/components/responses/InternalServerError' '503': $ref: '#/components/responses/ServiceUnavailable' components: schemas: ParaformerRequest: type: object required: - model - file_urls properties: model: description: |- `paraformer-v2`：支持中英日等多语种 `paraformer-8k-v2`：8kHz 采样率，仅支持中文 examples: - paraformer-v2 - paraformer-8k-v2 type: string default: paraformer-v2 file_urls: description: > 音频文件 URL 列表 **说明：** - 支持 HTTP/HTTPS 协议的公网可访问 URL - 单次请求最多 100 个 URL - 支持格式：aac, amr, avi, flac, flv, m4a, mkv, mov, mp3, mp4, mpeg, ogg, opus, wav, webm, wma, wmv - 单文件不超过 2GB，时长不超过 12 小时 items: type: string maxItems: 100 minItems: 1 type: array example: - https://example.com/audio/meeting.wav language_hints: anyOf: - items: type: string type: array - type: 'null' default: null description: > 待识别语种提示 **说明：** - 仅 `paraformer-v2` 支持，`paraformer-8k-v2` 不适用 - 支持的语言代码：`zh`（中文）、`en`（英文）、`ja`（日语）、`yue`（粤语）、`ko`（韩语）、`de`（德语）、`fr`（法语）、`ru`（俄语） x-unsupported-models: - paraformer-8k-v2 example: - zh - en channel_id: anyOf: - items: type: integer type: array - type: 'null' default: null description: |- 音轨索引 **说明：** - 索引从 0 开始，`[0]` 表示第一个音轨 - 默认 `[0]`（仅处理第一个音轨） - 每个指定的音轨独立计费 > 非必须不要传这个参数。 x-advanced: true example: - 0 recognition: anyOf: - description: Recognition configuration settings. properties: disfluency_removal_enabled: anyOf: - type: boolean - type: 'null' default: null description: | 过滤语气词 **说明：** - 启用后，识别结果中的语气词（如「嗯」「啊」）将被过滤 - 默认关闭 example: false timestamp_alignment_enabled: anyOf: - type: boolean - type: 'null' default: null description: | 时间戳校准 **说明：** - 启用后提高时间戳的对齐精度 - 默认关闭 example: false vocabulary_id: anyOf: - type: string - type: 'null' default: null description: | 热词表 ID **说明：** - 使用 DashScope 定制热词功能创建的热词表 ID - 仅 v2 系列模型支持 example: vocab-xxxx special_word_filter: anyOf: - type: string - type: 'null' default: null description: | 敏感词过滤规则（JSON 字符串） **说明：** - 不传则使用系统内置敏感词过滤 - 传入 JSON 可自定义替换/移除规则 example: >- {"filter_with_signed":{"word_list":["测试"]},"system_reserved_filter":true} type: object - type: 'null' default: null description: |- 识别配置 **说明：** - 包含语气词过滤、时间戳校准、热词、敏感词过滤等设置 - 不传则使用默认配置 > 非必须不要传这个参数。 x-advanced: true diarization: anyOf: - description: Speaker diarization configuration. properties: enabled: anyOf: - type: boolean - type: 'null' default: null description: | 说话人分离开关 **说明：** - 启用后识别结果中包含 `speaker_id` 字段 - 仅适用于单声道音频 - 默认关闭 example: false speaker_count: anyOf: - maximum: 100 minimum: 2 type: integer - type: 'null' default: null description: | 说话人数量提示 **说明：** - 范围：`2` - `100` - 需先启用说话人分离（`enabled: true`） - 仅作为辅助提示，不保证输出精确匹配该数量 example: 2 type: object - type: 'null' default: null description: |- 说话人分离配置 **说明：** - 包含分离开关和说话人数量提示 - 不传则不启用说话人分离 > 非必须不要传这个参数。 x-advanced: true TaskResponse: type: object properties: created: type: integer description: 任务创建时间戳 example: 1757165031 id: type: string description: 任务ID example: task-unified-1757165031-uyujaw3d model: type: string description: 实际使用的模型名称 object: type: string description: 任务的具体类型 enum: - audio.generation.task progress: type: integer description: 任务进度百分比 (0-100) minimum: 0 maximum: 100 example: 0 status: type: string description: 任务状态 enum: - pending - processing - completed - failed example: pending task_info: type: object description: 异步任务信息 properties: can_cancel: type: boolean description: 任务是否可以取消 example: true estimated_time: type: integer description: 预计完成时间（秒） example: 45 type: type: string description: 任务的输出类型 enum: - audio example: audio ErrorResponse400: type: object properties: error: type: object properties: message: type: string example: 请求格式错误 type: type: string example: invalid_request_error ErrorResponse401: type: object properties: error: type: object properties: message: type: string example: API密钥无效 type: type: string example: authentication_error ErrorResponse402: type: object properties: error: type: object properties: message: type: string example: 账户余额不足 type: type: string example: insufficient_quota ErrorResponse422: type: object properties: error: type: object properties: message: type: string example: 参数校验失败 type: type: string example: validation_error ErrorResponse429: type: object properties: error: type: object properties: message: type: string example: 请求频率超限 type: type: string example: rate_limit_error ErrorResponse500: type: object properties: error: type: object properties: message: type: string example: 服务器内部错误 type: type: string example: server_error ErrorResponse503: type: object properties: error: type: object properties: message: type: string example: 服务暂不可用，请稍后重试 type: type: string example: service_unavailable responses: TaskCreated: description: 任务创建成功 content: application/json: schema: $ref: '#/components/schemas/TaskResponse' BadRequest: description: 请求格式错误 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse400' Unauthorized: description: 未授权 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse401' PaymentRequired: description: 余额不足 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse402' UnprocessableEntity: description: 参数校验失败 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse422' TooManyRequests: description: 请求频率超限 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse429' InternalServerError: description: 服务器内部错误 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse500' ServiceUnavailable: description: 服务暂不可用 content: application/json: schema: $ref: '#/components/schemas/ErrorResponse503' securitySchemes: BearerAuth: type: http scheme: bearer description: | ## 所有接口均需要使用Bearer Token进行认证 ## 使用时在请求头中添加： `Authorization: Bearer YOUR_API_KEY` ````