创建语音

POST/v1/audio/speech

将文本转换为语音音频。当所选模型支持时，该端点支持纯文本生成、参考音频生成和参考图片生成。

认证

Authorization Bearer

在 Authorization 请求头中使用 API Key 作为 Bearer Token。

Request Body

inputstringrequired

Text input to synthesize into speech.

modelstringrequired

Model ID.

language_typestring

Speech language type.

speednumber

Speech speed adjustment.

referencesarray

Speech synthesis reference resources.

references.audio_urlstring

Reference audio URL.

references.image_urlstring

Reference image URL.

references.speakerstring

Provider speaker ID or cloned voice ID.

references.audio_datastring

Base64-encoded reference audio data.

references.image_datastring

Base64-encoded reference image data.

audio_configobject

Output audio configuration.

audio_config.formatstring

Output audio format.

audio_config.sample_rateinteger

Output audio sample rate in Hz.

audio_config.speech_ratenumber

Speech speed adjustment.

audio_config.pitch_ratenumber

Output pitch adjustment.

audio_config.loudness_ratenumber

Output loudness adjustment.

audio_formatstring

Output audio format.

rateinteger

Output audio sample rate in Hz.

loudness_rateinteger

Output loudness adjustment.

pitch_rateinteger

Output pitch adjustment.

watermarkobject

Watermark configuration.

watermark.aigc_watermarkboolean

Enable or disable the AIGC watermark.

watermark.aigc_metadataobject

AIGC metadata configuration.

response_formatstring

Response format.

metadataobject

Application metadata.

extra_bodyobject

Additional request body fields.

provider_optionsobject

Upstream configuration.

x_api_request_idstring

Upstream request ID.

Response

audiostring

生成音频 URL；如果 URL 转换不可用，则可能为 Base64 编码音频。

audio_typestring

audio 字段的编码方式。

modelstring

用于合成的模型。

voicestring

可用时返回供应商音色标识。纯文本请求可能返回默认音色值。

formatstring

音频格式。

sample_rateinteger

采样率，单位 Hz。

usageobject

用量和供应商侧关联信息。

usage.audio_duration_secondsnumber

可用时用于计费的生成音频秒数。优先使用供应商侧时长元数据，缺失时回退到后处理时长。

usage.original_durationnumber

供应商原始输出时长，单位秒。

usage.durationnumber

后处理后的音频时长，单位秒。

usage.provider_request_idstring

用于上游关联排查的供应商请求 ID。

请求

1curl -X POST https://api.token360.ai/v1/audio/speech \
2  -H "Authorization: Bearer sk-your-api-key" \
3  -H "Content-Type: application/json" \
4  -H "X-Api-Request-Id: seed-audio-text-only-001" \
5  -d '{
6    "model": "seed-audio-1.0",
7    "input": "Welcome to Token360, your unified AI gateway.",
8    "audio_format": "mp3",
9    "rate": 44100,
10    "speed": 0,
11    "loudness_rate": 0,
12    "pitch_rate": 0,
13    "response_format": "url"
14  }'

响应

JSON

1{
2  "audio": "your-audio-url",
3  "audio_type": "url",
4  "model": "seed-audio-1.0",
5  "voice": "default",
6  "format": "mp3",
7  "sample_rate": 44100,
8  "usage": {
9    "duration": 2.1,
10    "original_duration": 2.1,
11    "audio_duration_seconds": 2.1,
12    "provider_request_id": "seed-audio-text-only-001"
13  }
14}

此页面对您有帮助吗？