@crafter/trx 0.1.0 → 0.2.0

package/bin/trx.ts

@@ -10,7 +10,7 @@ const program = new Command();
 program
 	.name("trx")
 	.description("Agent-first CLI for audio/video transcription via Whisper")
-	.version("0.1.0")
+	.version("0.2.0")
 	.option("-o, --output <format>", "output format (json, table, auto)", "auto")
 	.hook("preAction", (thisCommand) => {
 		const opts = thisCommand.opts();

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@crafter/trx",
-	"version": "0.1.0",
+	"version": "0.2.0",
 	"description": "Agent-first CLI for audio/video transcription via Whisper",
 	"module": "bin/trx.ts",
 	"type": "module",
@@ -38,5 +38,8 @@
 	},
 	"devDependencies": {
 		"@types/bun": "latest"
+	},
+	"engines": {
+		"bun": ">=1.0.0"
 	}
 }

package/src/commands/transcribe.ts

@@ -105,9 +105,12 @@ export function createTranscribeCommand(): Command {
 					onStep: (step) => {
 						if (spinner) spinner.start(step);
 					},
+					onProgress: (progress) => {
+						if (spinner) spinner.message(`Transcribing... ${progress.percent}%`);
+					},
 				});
 
-				if (spinner) spinner.stop("Done");
+				if (spinner) spinner.stop("Transcription complete");
 
 				const filtered = opts.fields ? filterFields(result, opts.fields) : result;
 				output(format, {
@@ -118,12 +121,16 @@ export function createTranscribeCommand(): Command {
 							["Input", result.input],
 							["Language", result.metadata.language],
 							["Model", result.metadata.model],
-							["SRT", result.files.srt],
 							["TXT", result.files.txt],
-							["WAV", result.files.wav],
+							["SRT", result.files.srt],
 						],
 					},
 				});
+
+				if (isTTY) {
+					const wordCount = result.text.split(/\s+/).filter(Boolean).length;
+					p.note(`${wordCount} words transcribed\n\nopen ${result.files.txt}`, "Next");
+				}
 			} catch (e) {
 				outputError((e as Error).message, format);
 			}

package/src/core/pipeline.ts

@@ -2,7 +2,7 @@ import { basename, resolve } from "node:path";
 import type { TrxConfig } from "../utils/config.ts";
 import { cleanAudio } from "./audio.ts";
 import { downloadMedia } from "./download.ts";
-import { transcribe } from "./whisper.ts";
+import { type WhisperProgress, transcribe } from "./whisper.ts";
 
 export interface PipelineOptions {
 	input: string;
@@ -13,6 +13,7 @@ export interface PipelineOptions {
 	noDownload?: boolean;
 	noClean?: boolean;
 	onStep?: (step: string) => void;
+	onProgress?: (progress: WhisperProgress) => void;
 }
 
 export interface PipelineResult {
@@ -52,7 +53,7 @@ export async function runPipeline(opts: PipelineOptions): Promise<PipelineResult
 
 	const whisperInput = opts.noClean ? inputFile : wavPath;
 	opts.onStep?.("Transcribing with Whisper...");
-	const result = await transcribe(whisperInput, config, opts.language);
+	const result = await transcribe(whisperInput, config, opts.language, opts.onProgress);
 
 	return {
 		success: true,

package/src/core/whisper.ts

@@ -1,6 +1,10 @@
 import { existsSync, readFileSync } from "node:fs";
 import type { TrxConfig } from "../utils/config.ts";
-import { spawnOrThrow } from "../utils/spawn.ts";
+import { spawnOrThrow, spawnStreaming } from "../utils/spawn.ts";
+
+export interface WhisperProgress {
+	percent: number;
+}
 
 export interface WhisperResult {
 	srtPath: string;
@@ -12,6 +16,7 @@ export async function transcribe(
 	wavPath: string,
 	config: TrxConfig,
 	languageOverride?: string,
+	onProgress?: (progress: WhisperProgress) => void,
 ): Promise<WhisperResult> {
 	if (!existsSync(config.modelPath)) {
 		throw new Error(`Whisper model not found: ${config.modelPath}\nRun "trx init" to download a model.`);
@@ -42,7 +47,17 @@ export async function transcribe(
 	args.push("--entropy-thold", String(flags.entropyThold));
 	args.push("--logprob-thold", String(flags.logprobThold));
 
-	await spawnOrThrow(args, "whisper-cli transcription");
+	if (onProgress) {
+		args.push("--print-progress");
+		await spawnStreaming(args, "whisper-cli transcription", (line) => {
+			const match = line.match(/progress\s*=\s*(\d+)%/i);
+			if (match) {
+				onProgress({ percent: Number.parseInt(match[1], 10) });
+			}
+		});
+	} else {
+		await spawnOrThrow(args, "whisper-cli transcription");
+	}
 
 	const srtPath = `${wavPath}.srt`;
 	if (!existsSync(srtPath)) {
@@ -62,6 +77,7 @@ function srtToPlainText(srt: string): string {
 	return srt
 		.split("\n")
 		.filter((line) => !/^\[|-->/.test(line))
+		.filter((line) => !/^\d+\s*$/.test(line))
 		.filter((line) => line.trim().length > 0)
 		.join("\n");
 }

package/src/utils/spawn.ts

@@ -25,3 +25,40 @@ export async function spawnOrThrow(cmd: string[], context: string): Promise<stri
 	}
 	return result.stdout;
 }
+
+export async function spawnStreaming(
+	cmd: string[],
+	context: string,
+	onStderr?: (line: string) => void,
+): Promise<string> {
+	const proc = Bun.spawn(cmd, {
+		stdout: "pipe",
+		stderr: "pipe",
+	});
+
+	const stderrReader = (async () => {
+		const reader = proc.stderr.getReader();
+		const decoder = new TextDecoder();
+		let buffer = "";
+		while (true) {
+			const { done, value } = await reader.read();
+			if (done) break;
+			buffer += decoder.decode(value, { stream: true });
+			const lines = buffer.split("\n");
+			buffer = lines.pop() || "";
+			for (const line of lines) {
+				if (line.trim()) onStderr?.(line.trim());
+			}
+		}
+		if (buffer.trim()) onStderr?.(buffer.trim());
+	})();
+
+	const stdout = await new Response(proc.stdout).text();
+	await stderrReader;
+	const exitCode = await proc.exited;
+
+	if (exitCode !== 0) {
+		throw new Error(`${context} failed (exit ${exitCode})`);
+	}
+	return stdout.trim();
+}