Added 60dB integration#218
Open
manishEMS47 wants to merge 1 commit into
Open
Conversation
Contributor
📝 WalkthroughWalkthroughthis pr adds 60db as a speech provider across the sdk. you get tts/stt REST calls wrapped in a provider class, ai sdk v3 integration with optional alignment recovery, model definitions for billing, provider selection in voice actions, and full test/doc coverage. Changes60db Text-to-Speech Provider Integration
Possibly related PRs
Poem
🎯 4 (Complex) | ⏱️ ~45 minutes 🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (2 warnings)
✅ Passed checks (3 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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
src/ai-sdk/providers/60db.ts (1)
174-176: ⚡ Quick winunnecessary json round-trip
you're doing
JSON.parse(JSON.stringify({ sixtydb: { alignment } }))but can just assign the object directly. the alignment is already a plain object with plain arrays, so it's already json-serializable.skip the round-trip meow
if (words.length > 0) { const alignment = wordsToCharacterAlignment(words); - providerMetadata = JSON.parse( - JSON.stringify({ sixtydb: { alignment } }), - ); + providerMetadata = { sixtydb: { alignment } }; }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/ai-sdk/providers/60db.ts` around lines 174 - 176, The assignment to providerMetadata uses an unnecessary JSON round-trip; replace providerMetadata = JSON.parse(JSON.stringify({ sixtydb: { alignment } })) with a direct object assignment like providerMetadata = { sixtydb: { alignment } } so the plain JS object is used without serializing/deserializing; update the code path that sets providerMetadata (the variable named providerMetadata in the sixtydb provider code where alignment is referenced) accordingly.src/providers/60db.ts (1)
100-100: 💤 Low valuejob id prefix doesn't match provider name
you're using
db60_but the provider name is60db. might wanna keep them consistent for easier debugging later.tiny consistency fix
- const jobId = `db60_${Date.now()}_${Math.random().toString(36).slice(2)}`; + const jobId = `60db_${Date.now()}_${Math.random().toString(36).slice(2)}`;🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/providers/60db.ts` at line 100, The jobId prefix is inconsistent with the provider name: update the job id construction (the const jobId assignment that currently uses `db60_`) to use the provider name prefix `60db_` instead so identifiers match the provider; locate the `jobId` variable in this file (the template string using Date.now() and Math.random()) and replace the `db60_` prefix with `60db_` (and update any related places that rely on that exact prefix if present).
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/definitions/models/60db.ts`:
- Line 14: The schema's text field documents a 5000-char cap but doesn't enforce
it—update the zod schema for the text property (the line defining text:
z.string().describe(...)) to include .max(5000) (and an appropriate error
message); separately, clamp the computed pricing to the declared bounds by
ensuring any function that returns minUsd/maxUsd (the pricing computation in
this module) enforces minUsd <= price <= maxUsd before returning (or returns the
clamped minUsd/maxUsd values), and apply the same fix to the related pricing
code referenced around the minUsd/maxUsd fields.
In `@src/providers/60db.ts`:
- Line 13: The code imports writeFileSync from "node:fs" and uses it to write
the TTS output; replace that synchronous node fs usage with Bun.file to make
textToSpeech fully async: remove the writeFileSync import, find the textToSpeech
function (and the line where writeFileSync is called) and use
Bun.file(...).write or .writeSync as appropriate (prefer .write which returns a
Promise) to write the speech buffer/bytes, ensuring you await the write so
textToSpeech remains async and returns only after the file is persisted.
---
Nitpick comments:
In `@src/ai-sdk/providers/60db.ts`:
- Around line 174-176: The assignment to providerMetadata uses an unnecessary
JSON round-trip; replace providerMetadata = JSON.parse(JSON.stringify({ sixtydb:
{ alignment } })) with a direct object assignment like providerMetadata = {
sixtydb: { alignment } } so the plain JS object is used without
serializing/deserializing; update the code path that sets providerMetadata (the
variable named providerMetadata in the sixtydb provider code where alignment is
referenced) accordingly.
In `@src/providers/60db.ts`:
- Line 100: The jobId prefix is inconsistent with the provider name: update the
job id construction (the const jobId assignment that currently uses `db60_`) to
use the provider name prefix `60db_` instead so identifiers match the provider;
locate the `jobId` variable in this file (the template string using Date.now()
and Math.random()) and replace the `db60_` prefix with `60db_` (and update any
related places that rely on that exact prefix if present).
🪄 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: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 9b84d15f-373c-425b-9c9f-7117730e79e5
📒 Files selected for processing (18)
.claude/rules/video-generation.md.claude/settings.local.json.claude/skills/ffmpeg-mixing/SKILL.md.claude/skills/varg-video-generation/GUIDE.md.claude/skills/varg-video-generation/SKILL.md.env.exampleREADME.mdsrc/ai-sdk/index.tssrc/ai-sdk/providers/60db.tssrc/core/schema/shared.tssrc/definitions/actions/voice.tssrc/definitions/models/60db.tssrc/definitions/models/index.tssrc/providers/60db.tssrc/providers/index.tssrc/react/resolve.tssrc/tests/all.test.tssrc/tests/unit.test.ts
💤 Files with no reviewable changes (5)
- .claude/skills/varg-video-generation/SKILL.md
- .claude/rules/video-generation.md
- .claude/settings.local.json
- .claude/skills/varg-video-generation/GUIDE.md
- .claude/skills/ffmpeg-mixing/SKILL.md
|
|
||
| // Input schema with Zod | ||
| const sixtydbInputSchema = z.object({ | ||
| text: z.string().describe("Text to convert to speech (max 5000 characters)"), |
Contributor
There was a problem hiding this comment.
enforce the declared 5k text cap and pricing bounds in code
line 14 documents a 5000-char cap but doesn’t validate it, and line 64 can return values outside the declared minUsd/maxUsd range. that makes pricing/bounds inconsistent for billing consumers.
suggested patch
const sixtydbInputSchema = z.object({
- text: z.string().describe("Text to convert to speech (max 5000 characters)"),
+ text: z
+ .string()
+ .min(1)
+ .max(5000)
+ .describe("Text to convert to speech (max 5000 characters)"),
voice_id: z.string().optional().describe("60db voice_id (UUID) to use"),
@@
pricing: {
"60db": {
description: "$0.00002 per character via 60db ($0.01 minimum).",
- calculate: ({ characters = 500 }) => 0.00002 * characters,
+ calculate: ({ characters = 500 }) => {
+ const boundedCharacters = Math.min(Math.max(characters, 0), 5000);
+ const raw = 0.00002 * boundedCharacters;
+ return Math.min(0.1, Math.max(0.01, raw));
+ },
minUsd: 0.01,
maxUsd: 0.1, // ~5,000 char max input
},
},Also applies to: 63-66
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/definitions/models/60db.ts` at line 14, The schema's text field documents
a 5000-char cap but doesn't enforce it—update the zod schema for the text
property (the line defining text: z.string().describe(...)) to include
.max(5000) (and an appropriate error message); separately, clamp the computed
pricing to the declared bounds by ensuring any function that returns
minUsd/maxUsd (the pricing computation in this module) enforces minUsd <= price
<= maxUsd before returning (or returns the clamped minUsd/maxUsd values), and
apply the same fix to the related pricing code referenced around the
minUsd/maxUsd fields.
| * speech providers are interchangeable behind the `voice` action. | ||
| */ | ||
|
|
||
| import { writeFileSync } from "node:fs"; |
Contributor
There was a problem hiding this comment.
use bun.file instead of node:fs
the coding guidelines say to prefer Bun.file over node:fs readFile/writeFile. you're using this on line 200 for writing the output file.
✨ switch to bun.file api
-import { writeFileSync } from "node:fs";then on line 200:
if (outputPath) {
- writeFileSync(outputPath, buffer);
+ await Bun.write(outputPath, buffer);
console.log(`[60db] saved to ${outputPath}`);
}note: this makes textToSpeech fully async, which it already is meow
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/providers/60db.ts` at line 13, The code imports writeFileSync from
"node:fs" and uses it to write the TTS output; replace that synchronous node fs
usage with Bun.file to make textToSpeech fully async: remove the writeFileSync
import, find the textToSpeech function (and the line where writeFileSync is
called) and use Bun.file(...).write or .writeSync as appropriate (prefer .write
which returns a Promise) to write the speech buffer/bytes, ensuring you await
the write so textToSpeech remains async and returns only after the file is
persisted.
Source: Coding guidelines
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.
What
Brief description of what this PR does.
Why
Why is this change needed?
How to test
Steps to verify the changes work correctly.
Checklist
bun run check