tckm.yan/aha
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #1 from jhqxxx/main

Browse Source 
sync upstream
This commit is contained in:
Wei Cai
2026-03-12 12:02:11 +08:00
committed by GitHub
parent 1905aecae0 8ec2ef6ad3
commit 83d6c0b480
4 changed files with 278 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.gitignore
+4
View File
@@ -17,3 +17,7 @@ docs/MODEL_DEVELOPMENT_GUIDE.md
bak
.gitignore
demos
benchmarks
docs/api.md
+125
View File
@@ -277,6 +277,9 @@ curl http://127.0.0.1:10100/chat/completions \
}'
```
> **Note:** For OpenAI-standard audio transcription with `multipart/form-data` file upload,
> see the [Audio Transcriptions](#audio-transcriptions) endpoint.
**Streaming Response:**
```bash
@@ -376,6 +379,128 @@ Returns audio data in base64 WAV format.
- `voxcpm`, `voxcpm1.5`
### Audio Transcriptions
Transcribe audio files to text (Automatic Speech Recognition).
This endpoint provides OpenAI-compatible audio transcription using `multipart/form-data` format.
#### Endpoints
```
POST /audio/transcriptions
POST /v1/audio/transcriptions
```
Both endpoints use the same handler and return identical responses. The `/v1/audio/transcriptions` path follows OpenAI's standard API convention.
#### Request Body
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | file | Yes | The audio file to transcribe (wav, mp3, m4a, etc.) |
| `model` | string | No | Model identifier (optional, ignored - uses loaded model) |
| `language` | string | No | Language code (e.g., "zh", "en", "yue") |
| `prompt` | string | No | Optional text to guide transcription (not implemented, ignored) |
| `response_format` | string | No | Response format, only "json" or "text" supported (default: "json") |
| `temperature` | number | No | Sampling temperature (0.0 to 1.0, default: 0.0) |
#### Supported Languages
| Code | Language | Code | Language |
|------|----------|------|----------|
| `zh` | Chinese | `en` | English |
| `yue` | Cantonese | `ar` | Arabic |
| `de` | German | `fr` | French |
| `es` | Spanish | `pt` | Portuguese |
| `id` | Indonesian | `it` | Italian |
| `ko` | Korean | `ru` | Russian |
| `th` | Thai | `vi` | Vietnamese |
| `ja` | Japanese | `tr` | Turkish |
| `hi` | Hindi | `ms` | Malay |
| `nl` | Dutch | `sv` | Swedish |
| `da` | Danish | `fi` | Finnish |
| `pl` | Polish | `cs` | Czech |
| `fil` | Filipino | `fa` | Persian |
| `el` | Greek | `ro` | Romanian |
| `hu` | Hungarian | `mk` | Macedonian |
#### Examples
**Basic transcription:**
```bash
curl -X POST http://127.0.0.1:10100/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./audio.wav" \
-F model="qwen3asr-0.6b"
```
**With language specification:**
```bash
curl -X POST http://127.0.0.1:10100/v1/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./chinese_audio.wav" \
-F model="qwen3asr-0.6b" \
-F language="zh"
```
**With temperature:**
```bash
curl -X POST http://127.0.0.1:10100/v1/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./audio.wav" \
-F model="qwen3asr-0.6b" \
-F temperature="0.0"
```
#### Response
**Success (HTTP 200):**
```json
{
"text": "Transcribed text from the audio file"
}
```
**Error (HTTP 400):**
```json
{
"error": {
"message": "Audio file is required",
"type": "invalid_request_error",
"code": "missing_file"
}
}
```
**Error (HTTP 503):**
```json
{
"error": {
"message": "Model not initialized",
"type": "service_unavailable",
"code": "model_not_loaded"
}
}
```
#### Supported Models
- `qwen3asr-0.6b`
- `qwen3asr-1.7b`
- `glm-asr-nano-2512`
- `fun-asr-nano-2512`
#### File Upload Limit
Maximum audio file size: 100 MB
### Images Remove Background
Remove background from images.
docs/api.zh-CN.md
+125
View File
@@ -277,6 +277,9 @@ curl http://127.0.0.1:10100/chat/completions \
}'
```
> **提示**：支持 OpenAI 标准的音频转录接口，使用 `multipart/form-data` 上传文件。
> 详见 [语音转写](#语音转写) 章节。
**流式响应：**
```bash
@@ -376,6 +379,128 @@ curl http://127.0.0.1:10100/audio/speech \
- `voxcpm`、`voxcpm1.5`
### 语音转写
将音频文件转录为文本（自动语音识别）。
此接口提供符合 OpenAI 标准的音频转录功能，使用 `multipart/form-data` 格式。
#### 端点
```
POST /audio/transcriptions
POST /v1/audio/transcriptions
```
两个端点使用相同的处理函数并返回相同的响应。`/v1/audio/transcriptions` 路径遵循 OpenAI 的标准 API 约定。
#### 请求参数
| 参数 | 类型 | 必需 | 描述 |
|------|------|------|------|
| `file` | file | 是 | 要转录的音频文件 (wav, mp3, m4a 等) |
| `model` | string | 否 | 模型标识符（可选，被忽略 - 使用已加载的模型） |
| `language` | string | 否 | 语言代码（如 "zh"、"en"、"yue"） |
| `prompt` | string | 否 | 引导转录的可选文本（未实现，被忽略） |
| `response_format` | string | 否 | 响应格式，仅支持 "json" 或 "text"（默认："json"） |
| `temperature` | number | 否 | 采样温度 (0.0 到 1.0，默认：0.0) |
#### 支持的语言
| 代码 | 语言 | 代码 | 语言 |
|------|------|------|------|
| `zh` | 中文 | `en` | 英语 |
| `yue` | 粤语 | `ar` | 阿拉伯语 |
| `de` | 德语 | `fr` | 法语 |
| `es` | 西班牙语 | `pt` | 葡萄牙语 |
| `id` | 印尼语 | `it` | 意大利语 |
| `ko` | 韩语 | `ru` | 俄语 |
| `th` | 泰语 | `vi` | 越南语 |
| `ja` | 日语 | `tr` | 土耳其语 |
| `hi` | 印地语 | `ms` | 马来语 |
| `nl` | 荷兰语 | `sv` | 瑞典语 |
| `da` | 丹麦语 | `fi` | 芬兰语 |
| `pl` | 波兰语 | `cs` | 捷克语 |
| `fil` | 菲律宾语 | `fa` | 波斯语 |
| `el` | 希腊语 | `ro` | 罗马尼亚语 |
| `hu` | 匈牙利语 | `mk` | 马其顿语 |
#### 示例
**基本转录：**
```bash
curl -X POST http://127.0.0.1:10100/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./audio.wav" \
-F model="qwen3asr-0.6b"
```
**指定语言：**
```bash
curl -X POST http://127.0.0.1:10100/v1/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./chinese_audio.wav" \
-F model="qwen3asr-0.6b" \
-F language="zh"
```
**设置温度参数：**
```bash
curl -X POST http://127.0.0.1:10100/v1/audio/transcriptions \
-H "Authorization: Bearer NO_NEED" \
-F file="@./audio.wav" \
-F model="qwen3asr-0.6b" \
-F temperature="0.0"
```
#### 响应
**成功 (HTTP 200):**
```json
{
"text": "从音频文件转录的文本"
}
```
**错误 (HTTP 400):**
```json
{
"error": {
"message": "Audio file is required",
"type": "invalid_request_error",
"code": "missing_file"
}
}
```
**错误 (HTTP 503):**
```json
{
"error": {
"message": "Model not initialized",
"type": "service_unavailable",
"code": "model_not_loaded"
}
}
```
#### 支持的模型
- `qwen3asr-0.6b`
- `qwen3asr-1.7b`
- `glm-asr-nano-2512`
- `fun-asr-nano-2512`
#### 文件上传限制
最大音频文件大小：100 MB
### 图像背景移除
从图像中移除背景。
src/main.rs
+24 -2
View File
@@ -185,6 +185,18 @@ fn get_default_weight_path(model: WhichModel) -> String {
format!("{}/{}", save_dir, model_id)
}
/// Check if a model is downloaded by verifying the model directory exists
/// Returns true if ~/.aha/{model_id} directory exists, false otherwise
fn is_model_downloaded(model: WhichModel) -> bool {
let model_id = model.model_id();
let save_dir = match get_default_save_dir() {
Some(dir) => dir,
None => return false,
};
let model_path = format!("{}/{}", save_dir, model_id);
std::path::Path::new(&model_path).exists()
}
/// Model information for JSON output
#[derive(Serialize)]
struct ModelInfo {
@@ -192,6 +204,7 @@ struct ModelInfo {
model_id: String,
#[serde(rename = "type")]
model_type: String,
downloaded: bool,
}
/// List all supported models
@@ -232,6 +245,7 @@ fn run_list(args: ListArgs) -> anyhow::Result<()> {
name: possible_value.get_name().to_string(),
model_id: model.model_id().to_string(),
model_type: model.model_type().to_string(),
downloaded: is_model_downloaded(*model),
}
})
.collect();
@@ -240,13 +254,21 @@ fn run_list(args: ListArgs) -> anyhow::Result<()> {
// Table output (default)
println!("Available models:");
println!();
println!("{:<30} ModelScope ID", "Model Name");
println!(
"{:<30} {:<40} {:<10}",
"Model Name", "ModelScope ID", "Download"
);
println!("{}", "-".repeat(80));
for model in models {
let possible_value = model.to_possible_value().unwrap();
let name = possible_value.get_name();
let id = model.model_id();
println!("{:<30} {}", name, id);
let download_status = if is_model_downloaded(model) {
" ✔"
} else {
""
};
println!("{:<30} {:<40} {:<10}", name, id, download_status);
}
}