fix(voice-button): stabilize speech flow and clarify replace mode#315
Open
SonyLeo wants to merge 6 commits intoopentiny:developfrom
Open
fix(voice-button): stabilize speech flow and clarify replace mode#315SonyLeo wants to merge 6 commits intoopentiny:developfrom
SonyLeo wants to merge 6 commits intoopentiny:developfrom
Conversation
WalkthroughRenames voice modes from Changes
Sequence Diagram(s)sequenceDiagram
actor User
participant Button as VoiceButton
participant Speech as SpeechRecognition
participant Editor as Editor/Document
rect rgba(255, 200, 100, 0.5)
Note over User,Editor: Legacy / append-on-final flow
User->>Button: Press record
Button->>Speech: start()
Speech-->>Button: onresult (interim)
Button->>Button: emit interim event
Speech-->>Button: onresult (final)
Button->>Editor: append final transcript (if autoInsert)
Button->>Editor: focus
end
rect rgba(100, 200, 255, 0.5)
Note over User,Editor: New autoReplace / accumulated flow
User->>Button: Press record
Button->>Speech: start() (autoReplace config)
Speech-->>Button: onresult (finals + interim)
Button->>Button: parseSpeechRecognitionResult()
Button->>Button: update finalizedTranscript
Button->>Editor: replace text at speechRange with interim (finals+partial)
Speech-->>Button: onresult (final)
Button->>Editor: replace text at speechRange with merged final transcript
Button->>Button: reset session
Button->>Editor: focus
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Contributor
|
✅ Preview build completed successfully!
|
Contributor
📦 Package Previewpnpm add https://pkg.pr.new/@opentiny/tiny-robot@dc9998c pnpm add https://pkg.pr.new/@opentiny/tiny-robot-kit@dc9998c pnpm add https://pkg.pr.new/@opentiny/tiny-robot-svgs@dc9998c commit: dc9998c |
… to append/replace
SonyLeo
force-pushed
the
docs/update-speech-config
branch
from
4c60cae to
855ae9e
Compare
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (1)
packages/components/src/sender-actions/voice-button/index.vue (1)
25-57: Consider simplifyingspeechRangestate.The
speechRange.tofield is tracked (line 54) but never read—the replacement always usesdocSizeas the end position. Sincefromis always0for autoReplace mode, you could simplify to just a boolean flag indicating whether a recording session has started.However, if this tracking is intentional for future features (e.g., partial replacement from cursor position), feel free to keep it.
♻️ Optional simplification
-const speechRange = ref<{ from: number; to: number } | null>(null) +const hasStartedRecording = ref(false) const resetSpeechRange = () => { - speechRange.value = null + hasStartedRecording.value = false } const insertTranscript = (transcript: string) => { // ... early returns ... // autoReplace 模式：替换整个输入框内容 - if (speechRange.value === null) { - speechRange.value = { - from: 0, - to: 0, - } - } + hasStartedRecording.value = true const docSize = editorInstance.state.doc.content.size - const tr = editorInstance.state.tr.insertText(transcript, speechRange.value.from, docSize) + const tr = editorInstance.state.tr.insertText(transcript, 0, docSize) editorInstance.view.dispatch(tr) - speechRange.value = { - from: speechRange.value.from, - to: speechRange.value.from + transcript.length, - } editorInstance.commands.focus('end') }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@packages/components/src/sender-actions/voice-button/index.vue` around lines 25 - 57, The speechRange.to field is never read in insertTranscript (autoReplace uses docSize), so simplify state by removing speechRange.to and either replace speechRange with a boolean like speechStarted or keep only speechRange.from; update insertTranscript to initialize and check that flag (e.g., speechStarted or speechRange.from === 0) and use speechRange.from (or 0) as the insert start, and update the single state field accordingly; ensure references to speechRange.value.to are removed and editorInstance, autoReplace, and insertTranscript behavior remain unchanged.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/src/components/sender.md`:
- Line 1: The file docs/src/components/sender.md contains a UTF-8 BOM (U+FEFF)
at the start; remove the BOM so the file begins with the first markdown
character and re-save the file as UTF-8 without BOM (e.g., strip the leading
U+FEFF or re-encode), ensuring consistency with speech.types.ts which was
corrected earlier.
In `@packages/components/src/sender-actions/voice-button/speech.types.ts`:
- Line 1: The file speech.types.ts contains a UTF‑8 BOM at the very start;
remove the leading BOM character so the file begins with the first TypeScript
token, re-save the file as UTF‑8 without BOM, and re-commit; if your editor adds
BOMs automatically, update its save settings (or run a one-time clean-up
command) to ensure speech.types.ts stays UTF‑8 without BOM.
---
Nitpick comments:
In `@packages/components/src/sender-actions/voice-button/index.vue`:
- Around line 25-57: The speechRange.to field is never read in insertTranscript
(autoReplace uses docSize), so simplify state by removing speechRange.to and
either replace speechRange with a boolean like speechStarted or keep only
speechRange.from; update insertTranscript to initialize and check that flag
(e.g., speechStarted or speechRange.from === 0) and use speechRange.from (or 0)
as the insert start, and update the single state field accordingly; ensure
references to speechRange.value.to are removed and editorInstance, autoReplace,
and insertTranscript behavior remain unchanged.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: cc096f6e-f1c4-4f6d-b278-1a831a223417
📒 Files selected for processing (4)
docs/demos/sender/voice-input.vuedocs/src/components/sender.mdpackages/components/src/sender-actions/voice-button/index.vuepackages/components/src/sender-actions/voice-button/speech.types.ts
…inuous recognition support
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/src/components/sender.md`:
- Around line 167-173: The markdown table for the `lang` examples is missing the
required blank line before the admonition terminator (`:::`), causing the
terminator to be parsed as a table row; edit the `sender.md` section containing
the `| 值 | 说明 |` table and insert a single empty line after the last table row
(after the `| `en-US` | 美式英语 |` row) before the `:::` terminator so the
admonition closes correctly.
In `@packages/components/src/sender-actions/voice-button/webSpeechHandler.ts`:
- Around line 151-156: The manual stop is clearing event handlers before the
SpeechRecognition final onresult can arrive; change the stop flow so you call
this.recognition.stop() first and allow the native onresult→onend sequence to
run (letting onend invoke cleanup())—do not call cleanup() or
resetSessionTranscript() before recognition.stop(); also remove the eager
callbacks.onEnd() invocation at useSpeechHandler.ts line where callbacks.onEnd()
is called (around useSpeechHandler.ts:99) so the final transcript
(finalizedTranscript) is accumulated by the native events and delivered before
cleanup resets handlers/state.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 00f621fb-799e-4808-bf33-16701a4df123
📒 Files selected for processing (5)
docs/demos/sender/voice-input.vuedocs/src/components/sender.mdpackages/components/src/sender-actions/voice-button/index.vuepackages/components/src/sender-actions/voice-button/speech.types.tspackages/components/src/sender-actions/voice-button/webSpeechHandler.ts
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
docs/src/components/sender.md (1)
162-173:⚠️ Potential issue | 🟡 MinorClose the
langtip block.The
:::tip lang 语言说明block never terminates, so everything after the table can be parsed inside the admonition. Add the missing closing:::after the last row.📝 Suggested markdown fix
:::tip lang 语言说明 `lang` 用于指定语音识别语言，建议显式传入，并与页面的 `html lang` 保持一致，避免页面语言和浏览器环境语言不一致时出现识别偏差。 常见取值示例： | 值 | 说明 | | --- | --- | | `en` | 英语 | | `zh` | 中文 | | `zh-CN` | 简体中文 | | `en-US` | 美式英语 | + +::: #### 自定义语音服务🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/src/components/sender.md` around lines 162 - 173, The tip admonition starting with ":::tip lang 语言说明" is not closed, causing the remainder of the document to be included inside the block; locate the ":::tip lang 语言说明" block in docs/src/components/sender.md (the block containing the language table) and add the missing closing marker ":::" immediately after the table's last row so the admonition terminates properly.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@packages/components/src/sender-actions/voice-button/index.vue`:
- Around line 62-64: The current speechOptions constant snapshots
props.speechConfig once and becomes stale; make it reactive by replacing the
plain object with a computed or a reactive wrapper that derives from
props.speechConfig (e.g., replace const speechOptions = {...props.speechConfig}
with a computed(() => ({...props.speechConfig})) or reactive copy), and add a
watcher on props.speechConfig to recreate the speech recognition handler (the
customHandler / recognition handler creation logic) whenever speechConfig
changes so the handler uses updated lang, continuous, interimResults, and
customHandler settings.
---
Duplicate comments:
In `@docs/src/components/sender.md`:
- Around line 162-173: The tip admonition starting with ":::tip lang 语言说明" is
not closed, causing the remainder of the document to be included inside the
block; locate the ":::tip lang 语言说明" block in docs/src/components/sender.md (the
block containing the language table) and add the missing closing marker ":::"
immediately after the table's last row so the admonition terminates properly.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 6c69696d-a33a-4695-96e5-6e20d44d5b43
📒 Files selected for processing (4)
docs/demos/sender/voice-input.vuedocs/src/components/sender.mdpackages/components/src/sender-actions/voice-button/index.vuepackages/components/src/sender-actions/voice-button/speech.types.ts
🚧 Files skipped from review as they are similar to previous changes (1)
- docs/demos/sender/voice-input.vue
SonyLeo
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
变更背景
本次改动主要聚焦
VoiceButton的语音识别流程和文档说明，目的是统一“追加 / 替换”两种输入体验，并修复语音会话中的几个边界问题。主要收敛的问题包括：
autoReplace的行为不够直观，开启后在部分场景下看起来仍像“追加”continuous场景下识别结果会话语义不清晰，容易和替换模式产生理解偏差speechConfig动态更新后，底层语音 handler 仍可能沿用旧配置lang配置的说明不够清楚主要改动
1. 收敛语音写回语义
autoReplace: false时，语音识别结果继续追加到输入框末尾autoReplace: true时，录音期间使用识别结果替换整个输入框内容continuous: true场景下，替换内容为“当前录音会话的累计识别结果”，而不是仅保留最后一句2. 优化 Web Speech 结果处理
SpeechRecognitionEvent3. 修复 stop / end 生命周期问题
stop()时不再提前清理事件处理器onend统一完成收尾，避免丢失最后一段 transcript4. 让 speechConfig 按最新配置生效
speechConfiglang、continuous、interimResults、customHandler等配置改为在下一次start()时按最新值生效5. 同步更新文档和示例
autoReplace与continuous组合使用时的行为说明speechConfig.lang的配置说明和常见取值示例VoiceButton/SpeechConfig相关文档注释影响说明
autoReplace: true的语义明确为“整框替换”continuous: true，输入框会随着当前录音会话的累计识别结果持续更新验证情况
pnpm.cmd -F @opentiny/tiny-robot buildpnpm.cmd -F tiny-robot-test buildpnpm.cmd -F tiny-robot-test test -- src/voice-button/index.spec.tspnpm.cmd -F docs buildSummary by CodeRabbit
New Features
Documentation
Bug Fixes