From 9761932b4c8aaf45838ea3b26ab6054380bd0e98 Mon Sep 17 00:00:00 2001 From: clh02467605 Date: Tue, 23 Jun 2026 14:50:09 +0800 Subject: [PATCH] feat(omni): add voice listing functionality and update voice options --- packages/cli/src/commands/omni/chat.ts | 56 ++++++++++++++++++- .../cli/src/commands/speech/synthesize.ts | 16 ++++-- packages/cli/src/urls.ts | 3 + packages/cli/tests/e2e/omni.e2e.test.ts | 9 +++ skills/bailian-cli/reference/omni.md | 33 ++++++----- skills/bailian-cli/reference/speech.md | 36 ++++++------ 6 files changed, 113 insertions(+), 40 deletions(-) diff --git a/packages/cli/src/commands/omni/chat.ts b/packages/cli/src/commands/omni/chat.ts index cf71ff9..b827915 100644 --- a/packages/cli/src/commands/omni/chat.ts +++ b/packages/cli/src/commands/omni/chat.ts @@ -16,12 +16,52 @@ import { type StreamChunk, isInteractive, resolveFileUrl, + resolveOutputDir, + resolveCredential, } from "bailian-cli-core"; import { promptText, failIfMissing } from "../../output/prompt.ts"; import { emitResult } from "../../output/output.ts"; -import { resolveOutputDir, resolveCredential } from "bailian-cli-core"; -const OMNI_VOICES = ["Chelsie", "Cherry", "Ethan", "Serena", "Sunny", "Tina"]; +interface VoiceEntry { + voice: string; + name: string; + desc: string; + lang: string; +} + +// qwen-omni 系统音色 +const OMNI_VOICES: VoiceEntry[] = [ + { voice: "Dylan", name: "北京-晓东", desc: "胡同少年", lang: "中文/北京" }, + { voice: "Kiki", name: "粤语-阿清", desc: "甜美港妹", lang: "中文/英文" }, + { voice: "Li", name: "南京-老李", desc: "南京大叔", lang: "中文/英文" }, + { voice: "Sunny", name: "四川-晴儿", desc: "甜飒川妹", lang: "中文" }, + { voice: "Elias", name: "墨讲师", desc: "学术讲师女", lang: "中文/英文" }, + { voice: "Nofish", name: "不吃鱼", desc: "南方口音男", lang: "中文/英文" }, + { voice: "Marcus", name: "陕西-秦川", desc: "陕北汉子", lang: "中文/英文" }, + { voice: "Eric", name: "四川-程川", desc: "成都大哥", lang: "中文/英文" }, + { voice: "Jada", name: "上海-阿珍", desc: "沪上阿姐", lang: "中文" }, + { voice: "Rocky", name: "粤语-阿强", desc: "幽默港仔", lang: "中文/英文" }, + { voice: "Cherry", name: "芊悦", desc: "阳光自然女", lang: "中文/英文" }, + { voice: "Roy", name: "闽南-阿杰", desc: "闽南哥仔", lang: "中文/英文" }, + { voice: "Jennifer", name: "詹妮弗", desc: "美剧大女主", lang: "中文/英文" }, + { voice: "Ryan", name: "甜茶", desc: "美剧张力男", lang: "中文/英文" }, + { voice: "Katerina", name: "卡捷琳娜", desc: "御姐深情女", lang: "中文/英文" }, + { voice: "Peter", name: "天津-李彼得", desc: "天津捧哏", lang: "中文/英文" }, + { voice: "Ethan", name: "晨煦", desc: "北方口音男", lang: "中文/英文" }, +]; + +function printVoiceList(): void { + const col = (s: string, w: number) => s.padEnd(w); + process.stdout.write("\nOmni output voices:\n"); + process.stdout.write( + `${col("VOICE ID", 12)} ${col("NAME", 14)} ${col("DESCRIPTION", 14)} LANGUAGE\n`, + ); + process.stdout.write(`${"-".repeat(12)} ${"-".repeat(14)} ${"-".repeat(14)} ${"-".repeat(12)}\n`); + for (const v of OMNI_VOICES) { + process.stdout.write(`${col(v.voice, 12)} ${col(v.name, 14)} ${col(v.desc, 14)} ${v.lang}\n`); + } + process.stdout.write(`\nTotal: ${OMNI_VOICES.length} voices\n`); +} /** * Extension to input audio format. @@ -110,7 +150,11 @@ export default defineCommand({ }, { flag: "--voice ", - description: `Output voice (default: Cherry). Options: ${OMNI_VOICES.join(", ")}`, + description: "Output voice ID (default: Cherry). Use --list-voices to see all options", + }, + { + flag: "--list-voices", + description: "List available output voices and exit", }, { flag: "--audio-format ", description: "Audio output format (default: wav)" }, { flag: "--audio-out ", description: "Save audio to file (default: auto-generate)" }, @@ -119,6 +163,7 @@ export default defineCommand({ { flag: "--temperature ", description: "Sampling temperature (0.0, 2.0]", type: "number" }, ], examples: [ + "bl omni --list-voices", 'bl omni --message "Hello, who are you?"', 'bl omni --message "Describe this image" --image ./photo.jpg', 'bl omni --message "What is this audio saying?" --audio https://example.com/audio.wav', @@ -129,6 +174,11 @@ export default defineCommand({ 'bl omni --message "Read this passage aloud" --audio-out greeting.wav', ], async run(config: Config, flags: GlobalFlags) { + if (flags.listVoices) { + printVoiceList(); + return; + } + // --- Parse messages --- let userMessages: string[] = []; if (flags.message) { diff --git a/packages/cli/src/commands/speech/synthesize.ts b/packages/cli/src/commands/speech/synthesize.ts index 23d9874..ce6aee2 100644 --- a/packages/cli/src/commands/speech/synthesize.ts +++ b/packages/cli/src/commands/speech/synthesize.ts @@ -20,12 +20,14 @@ import { DOCS_HOSTS, } from "bailian-cli-core"; -const COSYVOICE_CLONE_DESIGN_DOC = `${DOCS_HOSTS.cn}/cosyvoice-clone-design-api`; +import { VOICE_TTS_PAGE } from "../../urls.ts"; import { downloadFile } from "../../utils/download.ts"; import { runConcurrent, downloadParallel, getConcurrency } from "../../utils/concurrent.ts"; import { promptText, promptSelect, failIfMissing } from "../../output/prompt.ts"; import { emitResult, emitBare } from "../../output/output.ts"; +const COSYVOICE_CLONE_DESIGN_DOC = `${DOCS_HOSTS.cn}/cosyvoice-clone-design-api`; + interface VoiceEntry { voice: string; name: string; @@ -37,7 +39,7 @@ interface VoiceEntry { const COSYVOICE_V3_FLASH_VOICES: VoiceEntry[] = [ // 社交陪伴 { voice: "longanyang", name: "龙安洋", desc: "阳光大男孩", lang: "中文/英文" }, - { voice: "longanhuan", name: "龙安欢", desc: "欢脱元气女", lang: "中文/英文" }, + { voice: "longanhuan_v3", name: "龙安欢", desc: "欢脱元气女", lang: "中文/英文" }, { voice: "longantai_v3", name: "龙安台", desc: "嗲甜台湾女", lang: "中文/英文" }, { voice: "longhua_v3", name: "龙华", desc: "元气甜美女", lang: "中文/英文" }, { voice: "longcheng_v3", name: "龙橙", desc: "智慧青年男", lang: "中文/英文" }, @@ -121,12 +123,14 @@ function printVoiceList(model: string): void { const voices = MODEL_VOICES[model]; if (!voices) { process.stdout.write(`No built-in voice list available for model: ${model}\n`); + process.stdout.write(`Browse voices in the console: ${VOICE_TTS_PAGE}\n`); return; } if (voices.length === 0) { process.stdout.write(`Model ${model} has no system voices.\n`); process.stdout.write("Use clone or design voices created via the CosyVoice API.\n"); process.stdout.write(`See: ${COSYVOICE_CLONE_DESIGN_DOC}\n`); + process.stdout.write(`Browse voices in the console: ${VOICE_TTS_PAGE}\n`); return; } const col = (s: string, w: number) => s.padEnd(w); @@ -139,6 +143,7 @@ function printVoiceList(model: string): void { process.stdout.write(`${col(v.voice, 26)} ${col(v.name, 10)} ${col(v.desc, 16)} ${v.lang}\n`); } process.stdout.write(`\nTotal: ${voices.length} voices\n`); + process.stdout.write(`Preview and browse more voices in the console: \n${VOICE_TTS_PAGE}\n`); } export default defineCommand({ @@ -156,11 +161,12 @@ export default defineCommand({ { flag: "--voice ", description: - "Voice ID. Use --list-voices to see system voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID", + "Voice ID. Use --list-voices to see built-in voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID", }, { flag: "--list-voices", - description: "List available system voices for the selected model and exit", + description: + "List built-in system voices for the selected model and exit (console link shown in output)", }, { flag: "--format ", description: "Audio format: mp3, pcm, wav, opus (default: mp3)" }, { flag: "--sample-rate ", description: "Audio sample rate in Hz (e.g. 24000)" }, @@ -264,7 +270,7 @@ export default defineCommand({ const modelVoices = MODEL_VOICES[model]; if (modelVoices && modelVoices.length > 0) { throw new BailianError( - `--voice is required.\nRun the following to see available voices:\n bl speech synthesize --list-voices --model ${model}`, + `--voice is required.\nRun the following to see available voices:\n bl speech synthesize --list-voices --model ${model}\nBrowse more voices: ${VOICE_TTS_PAGE}`, ExitCode.USAGE, ); } else { diff --git a/packages/cli/src/urls.ts b/packages/cli/src/urls.ts index 3d21a29..5a21206 100644 --- a/packages/cli/src/urls.ts +++ b/packages/cli/src/urls.ts @@ -14,3 +14,6 @@ export const BAILIAN_CONSOLE = `${BAILIAN_CONSOLE_ROOT}/cn-beijing`; /** Direct deep link to API key management page. */ export const API_KEY_PAGE = `${BAILIAN_CONSOLE}/?tab=app#/api-key`; + +/** Voice TTS experience center — browse system and custom voices. */ +export const VOICE_TTS_PAGE = "https://help.aliyun.com/zh/model-studio/cosyvoice-voice-list"; diff --git a/packages/cli/tests/e2e/omni.e2e.test.ts b/packages/cli/tests/e2e/omni.e2e.test.ts index f0f2a36..c3d25a5 100644 --- a/packages/cli/tests/e2e/omni.e2e.test.ts +++ b/packages/cli/tests/e2e/omni.e2e.test.ts @@ -15,6 +15,15 @@ describe("e2e: omni", () => { expect(exitCode, stderr).toBe(0); expect(stderr).toMatch(/omni|--message|--audio|text-only/i); }); + + test("omni --list-voices 输出音色列表并退出", async () => { + const { stdout, stderr, exitCode } = await runCli(["omni", "--list-voices"]); + expect(exitCode, stderr).toBe(0); + expect(stdout).toMatch(/Omni output voices:/); + expect(stdout).toMatch(/Dylan/); + expect(stdout).toMatch(/Cherry/); + expect(stdout).toMatch(/Total: 17 voices/); + }); }); describe.skipIf(!isBailianE2EMediaEnabled() || !isDashScopeE2EReady())( diff --git a/skills/bailian-cli/reference/omni.md b/skills/bailian-cli/reference/omni.md index 5e12aa8..24f0503 100644 --- a/skills/bailian-cli/reference/omni.md +++ b/skills/bailian-cli/reference/omni.md @@ -23,23 +23,28 @@ Index: [index.md](index.md) #### Options -| Flag | Type | Required | Description | -| ---------------------- | ------- | -------- | ------------------------------------------------------------------------------------ | -| `--message ` | array | yes | Message text (repeatable, prefix role: to set role) | -| `--model ` | string | no | Model ID (default: qwen3.5-omni-plus) | -| `--system ` | string | no | System prompt | -| `--image ` | array | no | Image URL or local file (repeatable) | -| `--audio ` | array | no | Audio URL or local file (.wav/.mp3/.amr/.aac/.m4a/.ogg/.3gp/.3gpp) | -| `--video ` | array | no | Video file URL / local path, or comma-separated frame URLs | -| `--voice ` | string | no | Output voice (default: Cherry). Options: Chelsie, Cherry, Ethan, Serena, Sunny, Tina | -| `--audio-format ` | string | no | Audio output format (default: wav) | -| `--audio-out ` | string | no | Save audio to file (default: auto-generate) | -| `--text-only` | boolean | no | Output text only, no audio generation | -| `--max-tokens ` | number | no | Maximum tokens to generate | -| `--temperature ` | number | no | Sampling temperature (0.0, 2.0] | +| Flag | Type | Required | Description | +| ---------------------- | ------- | -------- | ----------------------------------------------------------------------- | +| `--message ` | array | yes | Message text (repeatable, prefix role: to set role) | +| `--model ` | string | no | Model ID (default: qwen3.5-omni-plus) | +| `--system ` | string | no | System prompt | +| `--image ` | array | no | Image URL or local file (repeatable) | +| `--audio ` | array | no | Audio URL or local file (.wav/.mp3/.amr/.aac/.m4a/.ogg/.3gp/.3gpp) | +| `--video ` | array | no | Video file URL / local path, or comma-separated frame URLs | +| `--voice ` | string | no | Output voice ID (default: Cherry). Use --list-voices to see all options | +| `--list-voices` | boolean | no | List available output voices and exit | +| `--audio-format ` | string | no | Audio output format (default: wav) | +| `--audio-out ` | string | no | Save audio to file (default: auto-generate) | +| `--text-only` | boolean | no | Output text only, no audio generation | +| `--max-tokens ` | number | no | Maximum tokens to generate | +| `--temperature ` | number | no | Sampling temperature (0.0, 2.0] | #### Examples +```bash +bl omni --list-voices +``` + ```bash bl omni --message "Hello, who are you?" ``` diff --git a/skills/bailian-cli/reference/speech.md b/skills/bailian-cli/reference/speech.md index f4c5dbc..dfa8dd2 100644 --- a/skills/bailian-cli/reference/speech.md +++ b/skills/bailian-cli/reference/speech.md @@ -77,24 +77,24 @@ bl speech recognize --url https://example.com/audio.mp3 --no-wait --quiet #### Options -| Flag | Type | Required | Description | -| ---------------------- | ------- | -------- | ----------------------------------------------------------------------------------------------------------------------- | -| `--text ` | string | yes | Text to synthesize into speech | -| `--text-file ` | string | no | Read text from a file instead of --text | -| `--model ` | string | no | Model ID (default: cosyvoice-v3-flash). System voices available for cosyvoice-v3-flash | -| `--voice ` | string | no | Voice ID. Use --list-voices to see system voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID | -| `--list-voices` | boolean | no | List available system voices for the selected model and exit | -| `--format ` | string | no | Audio format: mp3, pcm, wav, opus (default: mp3) | -| `--sample-rate ` | string | no | Audio sample rate in Hz (e.g. 24000) | -| `--volume ` | string | no | Volume 0-100 (default: 50) | -| `--rate ` | string | no | Speech rate 0.5-2.0 (default: 1.0) | -| `--pitch ` | string | no | Pitch multiplier 0.5-2.0 (default: 1.0) | -| `--seed ` | string | no | Random seed 0-65535 for reproducible synthesis | -| `--language ` | string | no | Language hint (e.g. zh, en, ja, ko, fr, de) | -| `--instruction ` | string | no | Natural language instruction to control speech style (e.g. "Use a gentle tone"） | -| `--enable-ssml` | boolean | no | Enable SSML markup parsing in input text | -| `--out ` | string | no | Save audio to file (default: auto-generate in temp dir) | -| `--stream` | boolean | no | Stream raw PCM audio to stdout (pipe to player) | +| Flag | Type | Required | Description | +| ---------------------- | ------- | -------- | ------------------------------------------------------------------------------------------------------------------------- | +| `--text ` | string | yes | Text to synthesize into speech | +| `--text-file ` | string | no | Read text from a file instead of --text | +| `--model ` | string | no | Model ID (default: cosyvoice-v3-flash). System voices available for cosyvoice-v3-flash | +| `--voice ` | string | no | Voice ID. Use --list-voices to see built-in voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID | +| `--list-voices` | boolean | no | List built-in system voices for the selected model and exit (console link shown in output) | +| `--format ` | string | no | Audio format: mp3, pcm, wav, opus (default: mp3) | +| `--sample-rate ` | string | no | Audio sample rate in Hz (e.g. 24000) | +| `--volume ` | string | no | Volume 0-100 (default: 50) | +| `--rate ` | string | no | Speech rate 0.5-2.0 (default: 1.0) | +| `--pitch ` | string | no | Pitch multiplier 0.5-2.0 (default: 1.0) | +| `--seed ` | string | no | Random seed 0-65535 for reproducible synthesis | +| `--language ` | string | no | Language hint (e.g. zh, en, ja, ko, fr, de) | +| `--instruction ` | string | no | Natural language instruction to control speech style (e.g. "Use a gentle tone"） | +| `--enable-ssml` | boolean | no | Enable SSML markup parsing in input text | +| `--out ` | string | no | Save audio to file (default: auto-generate in temp dir) | +| `--stream` | boolean | no | Stream raw PCM audio to stdout (pipe to player) | #### Examples